gamage - how the ACA will create perverse …...The Affordable Care Act’s Perverse Incentives Tax...

Electronic copy available at: http://ssrn.com/abstract=2067138Electronic copy available at: http://ssrn.com/abstract=2067138

The Affordable Care Act’s Perverse Incentives

Tax L. Rev. (forthcoming)

1

How the Affordable Care Act Will Create Perverse Incentives Harming Low and

Moderate Income Workers

By David Gamage1

(Tax Law Review, forthcoming)

The Affordable Care Act2 has been heralded as the signature achievement of the Obama Administration.3 Called “Obamacare” by some, the Affordable Care Act (or “ACA”) is the most extensive reform to the American healthcare system since the creation of Medicare and Medicaid in 1965.4 With respect to tax administration, the ACA is arguably the most significant attempt ever to reform social welfare policy through the tax code.5

1 Assistant Professor, University of California, Berkeley, School of Law (Boalt Hall).

This Essay was written while David Gamage was on teaching leave from Berkeley while serving as Special Counsel and Senior Stanley S. Surrey Fellow to the U.S. Department of the Treasury, Office of Tax Policy. The views expressed in this Essay are solely those of David Gamage; nothing discussed herein should be attributed in any way to the Treasury Department, the Obama Administration, or anyone other than David Gamage.

Many thanks to Susannah Camic, Amy Monahan, Shana Lavarreda, Shanna Lehrman, Deborah Schenk, Shruti Rana, the other organizers and participants at the 2011 NYU-UCLA Tax Policy Conference, and the Tax Law Review editorial staff. 2 The “Affordable Care Act” (or ACA) refers jointly to the Patient Protection and Affordable Care Act (or PPACA) – Pub. L. No. 111-148, 124 Stat. 119 (2010) – and the Health Care Education Reconciliation Act of 2010 – Pub. L. No. 111-152, 124 Stat. 1029 (2010). The “Affordable Care Act” or “ACA” is the Obama Administration’s preferred term for referring to these healthcare reform acts and is the term used in the regulations interpreting the acts. E.g., Prop. Treas. Reg. § 36B, 76 Fed. Reg. 50931, 50932 (Aug. 17, 2011). Consequently, I will also refer to these health care reform acts as the “ACA” throughout this Essay. 3 E.g., Pema Levy, How the Obama Administration is Jeopardizing Health Care Reform, THE NEW REPUBLIC, October, 3rd, 2011, (referring to the Affordable Care Act as the Obama Administration’s “signature policy achievement”), available at http://www.tnr.com/article/politics/95631/supreme-court-case-medicaid-california-affordable-care-act. 4 E.g., The Future of Nursing: Leading Change, Advancing Health Recommendations from the IOM/RWJF Initiative on the Future of Nursing, available at http://www.cinhc.org/wordpress/wp-content/uploads/2011/02/IOM-report-summary.pdf (“The ACA represents the broadest changes to the health care system since the 1965 creation of the Medicare and Medicaid programs and is expected to provide insurance coverage for an additional 32 million previously uninsured Americans.”). 5 See Edward A. Zelinsky, The Health-Related Tax Provisions of PPACA and HCERA: Contingent, Complex, Incremental and Lacking Cost Controls, CARDOZO LEGAL STUDIES RESEARCH PAPER NO. 301 (2010), available at http://ssrn.com/abstract=1633556 (summarizing the many tax provisions of the Affordable Care Act).




2

The ACA promises many improvements to American health care.6 While recognizing the importance of these improvements, this Essay focuses on how the ACA will also create perverse incentives harming low- and moderate-income workers.7 This Essay explains how the ACA will impose effective taxes with respect to low- and moderate-income workers, thereby reducing these workers’ employment opportunities and creating a number of other economic and social harms.8

When a law or regulation deters economic actors from the choices that they would have otherwise made, we can say that the law or regulation imposes “effective taxes” on those choices.9 For the most part then, “effective taxes” are essentially synonymous with “perverse incentives.”10 This Essay argues that – once key provisions of the ACA come into effect in 2014 – the ACA will impose effective taxes on a number of important decisions affecting low- and moderate-income Americans, including:

The ACA will deter low- and moderate-income taxpayers from accepting jobs with employers that offer “affordable” health insurance;

6 E.g., THE WHITE HOUSE, AFFORDABLE CARE ACT – ABOUT THE NEW LAW, available at http://www.whitehouse.gov/healthreform/healthcare-overview. 7 This Essay does not argue that the ACA will create more harm than benefit for low- and moderate-income Americans. In my assessment, the ACA’s positive improvements to the American health care system will more than compensate for its harms. Moreover – although analyzing these issues is beyond the scope of this Essay – my primary doubts about the ACA arise from concerns about whether the ACA will be effective in the face of Republicans’ attempts to obstruct its successful implementation. Nevertheless, it is important to understand the costs that the ACA will impose on low- and moderate-income workers, particularly because most of these costs are avoidable. See Part.IV infra. Explaining how these costs will arise and how they might be prevented is the purpose of this Essay. 8 As the phrase is used in this Essay, “low- and moderate-income workers” generally refers to workers with household incomes higher than 133 percent of the federal poverty line and lower than (at most) 400 percent of the federal poverty line. Workers with household incomes below 133 percent of the federal poverty line will generally qualify for Medicaid, such that most of this Essay’s analysis will not apply. For analysis regarding the high-end threshold for “low- and moderate-income workers”, and for charts showing how percentages of the federal poverty line relate to actual household incomes, see Part II.C infra. 9 The term “effective taxes” is used to contrast with explicit taxes or statutory taxes. Unlike the latter terms, “effective taxes” include all of the ways in which a law or regulation increases the price of one economic decision as compared to alternative economic decisions. 10 More precisely, effective taxes are equivalent to perverse incentives to the extent that the effective taxes lead to harmful changes in behavior. But unless there is some reason for deterring the choices economic actors would have made in the absence of effective taxes (e.g., externalities), then effective taxes create perverse incentives almost by definition, following the baseline assumption that it is generally undesirable for governments to alter the incentives of economic actors unless there is a good reason for doing so. I primarily use the term “effective taxes” in this Essay because I do not mean to imply that there is anything inherently harmful about the choices economic actors make as a result of the effective taxes. For instance, if a low-income taxpayer decides not to accept a job, this decision may sometimes be in the best interests of both the taxpayer and society. But if a law or regulation imposes effective taxes that result in a low-income taxpayer not accepting a job that the taxpayer would otherwise have accepted, then this change in behavior will generally harm society. The term “effective taxes” captures this distinction better than does the term “perverse incentives,” and I thus use the term effective taxes in those instances in which it is important to capture this distinction. In other instances, I use the two terms interchangeably.




3

The ACA will discourage many low- and moderate-income taxpayers from attempting to increase their household incomes;

The ACA will penalize many low- and moderate-income taxpayers who choose to marry, and will incentivize many low- and moderate-income taxpayers to divorce;

The ACA will dissuade employers from hiring low- and moderate-income taxpayers, and will encourage employers to reduce the salaries paid to some low- and moderate-income employees;

The ACA will prompt employers to shift some low- and moderate-income employees from full-time positions to part-time positions;

The ACA will tempt employers to implement a number of other costly strategies for circumventing the ACA’s employer mandates and penalties;

The ACA will induce employers to stop offering “affordable” health insurance to at least some low- and moderate-income employees, and – if this occurs to a significant enough degree – the budgetary cost of the ACA may greatly exceed the official projections issued by the Congressional Budget Office.

Tragically, these effective taxes could have been avoided. We ought perhaps to accept these effective taxes were they a necessary cost of achieving the ACA’s many positive goals. But the ACA could have been drafted to attain its desirable ends without creating most of the effective taxes analyzed by this Essay. Moreover, there is still hope of reforming the ACA so as to preserve its positive features while mitigating or eliminating the effective taxes that the ACA will create without further reform.

The source of most of these effective taxes is the mismatch that the ACA will create between the tax subsidies available for employer-sponsored health insurance and those available for the health insurance purchased by individuals. Most higher-income taxpayers will receive much larger tax subsidies if they are offered employer-sponsored health insurance, whereas most lower-income taxpayers will receive much larger tax subsidies if they are not offered “affordable” employer-sponsored health insurance. This mismatch in the available tax subsidies results because the ACA maintains most of the previously existing tax benefits for employer-sponsored health insurance (which primarily benefit higher-income taxpayers), whereas the new



4

tax subsidies that the ACA will create for health insurance purchased by individuals will primarily benefit lower-income taxpayers.11

If the ACA had instead transformed the previously existing tax benefits for employer-sponsored health insurance into refundable tax credits structured in a similar fashion to the ACA’s new tax subsidies, then most of the perverse incentives analyzed by this Essay would have been completely alleviated.12 Moreover, many commentators have already called for reforming the previously existing tax benefits for employer-sponsored health insurance in exactly this fashion, arguing that these tax benefits are regressive and that they encourage excess health care consumption.13 This Essay explains why failing to reform the previously existing tax benefits for employer-sponsored health insurance will create far more harm once key provisions of the ACA come into effect in 2014.

This Essay proceeds in four parts. Part I explains why employers provided health insurance prior to the ACA, and Part II explains how the ACA will alter employers’ incentives as to whether to offer health insurance. Understanding how the ACA will affect employers’ incentives is key to understanding the ACA’s effective taxes. Parts I and II are directed toward readers with some background in tax law and theory, but who may not have previously immersed themselves in health care or the details of the ACA.14 Readers who already have an in-depth understanding of the tax provisions of the ACA may wish to start reading with Part III.

Part III explains how the ACA will create effective taxes harming low- and moderate-income workers, and Part IV concludes by explaining how these effective taxes could be avoided

11 And taxpayers will be ineligible for these new subsidies if the taxpayers’ employers offer “affordable” health insurance. 12 The exception is the perverse incentives some low- and moderate-income taxpayers will face to avoid increasing their incomes. Unlike the other perverse incentives analyzed in this Essay, these perverse incentives would remain even if the previously existing tax benefits for employer-sponsored health insurance were transformed into refundable tax credits as this Essay recommends. 13 E.g., Jonathan Gruber, The Tax Exclusion for Employer-Sponsored Health Insurance, NBER WORKING PAPER

15766, at 3 (2010), available at http://www.nber.org/papers/w15766; Paul N. Van de Water, Limiting the Tax Exclusion for Employer-Sponsored Insurance Can Help Pay for Health Reform: Universal Coverage May Be Out of Reach Otherwise, June 4, 2009, CENTER ON BUDGET POLICIES AND PRIORITIES, available at http://www.cbpp.org/cms/index.cfm?fa=view&id=2832; URBAN INSTITUTE AND BROOKINGS INSTITUTION – TAX POLICY CENTER, The Tax Policy Briefing Book: A Citizens’ Guide for the 2008 Election and Beyond, at 8-12, April 9, 2008, available at http://www.taxpolicycenter.org/upload/Elements/II-5KEYELEMENTS_HealthInsuranceandHealthCare.final.pdf; Jason Roffenbender, Employer-Based Health Insurance: Why Congress Should Cap Tax Benefits Consistently, December 5, 2008, THE HERITAGE FOUNDATION, available at http://www.heritage.org/Research/Reports/2008/12/Employer-Based-Health-Insurance-Why-Congress-Should-Cap-Tax-Benefits-Consistently; Bradley W. Joondeph, Tax Policy and Health Care Reform: Rethinking the Tax Treatment of Employer-Sponsored Health Insurance, BRIGHAM YOUNG UNIVERSITY LAW REVIEW 1229, 2005. 14 However, this Essay makes no attempt to summarize all of the major provisions of the ACA or even all of the major tax provisions of the ACA. For a summary and explanation of the ACA, see, e.g., WHAT DOES HEALTH

REFORM MEAN FOR YOU? A CONSUMER’S GUIDE, NATIONAL CENTER FOR POLICY ANALYSIS (2010); A SUMMARY

OF THE HEALTH REFORM LAW, FAMILIES USA (2010).



5

by reforming the tax benefits available for employer-sponsored health insurance. This Essay ends with a call for action. The ACA is an impressive accomplishment, but further reform is urgently needed to prevent the ACA from harming low- and moderate-income workers through effective taxes beginning in 2014.

I) WHY EMPLOYERS PROVIDED HEALTH INSURANCE PRIOR TO THE ACA

Prior to the ACA, why did most Americans receive health insurance from their employers?15 One might alternatively ask why Americans did not generally receive food, cars, or movie tickets from their employers. As with health insurance, employees like receiving food, cars, and movie tickets, and often choose to spend their own money on these items when employers compensate them with cash wages. But each employee places a different relative value on food, cars, movie tickets, and similar goods, and employees thus generally prefer to receive cash wages and then to decide for themselves how much of each of these goods to purchase rather than having their employers involved in those consumption decisions.16

Of course, sometimes employers provide these and other in-kind fringe benefits because providing those benefits serves a business purpose other than compensating employees.17 For example, an employer might provide an employee with a car if the employer wants the employee to use that car for work-related travel. But outside of such non-compensation-motivated scenarios, employers should generally only provide in-kind benefits in lieu of cash wages either if: (a) the employer can provide the in-kind benefit at a lower price than what it would cost for the employee to purchase the benefit, or (b) if it is tax favorable for the employer to provide the in-kind benefit in lieu of cash wages.18

As the remainder of this Part explains, both of these factors motivated employers to provide health insurance in lieu of cash wages prior to the ACA. Due primarily to adverse selection and risk classification, employers were able to offer better quality health insurance at a 15 Amy Monahan, The Complex Relationship Between Taxes and Health Insurance, MINNESOTA LEGAL STUDIES

RESEARCH PAPER NO. 10-1, at 1 (2010), available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1531322; Allison Hoffman, Oil and Water: Mixing Individual Mandates, Fragmented Markets, and Health Reform, 36 AM. J. L. AND MED. 7, 18 (2010) (“The majority of privately insured Americans still obtain their health insurance coverage through an employer….”); Robert J. Mills, U.S. Census Bureau Health Insurance Coverage: 2000 (Sept. 2001), available at http:// www.census.gov/prod/2001pubs/p60-215.pdf. 16 To the extent it costs employers resources to provide health insurance, employers could instead transfer those funds directly to employees in the form of higher wages. Each employee could then choose how much of these wages to spend on health insurance. By spending resources to subsidize health insurance rather than on wages, employers thus limit their employees’ options for how to use the fruits of their labor. 17 For a general discussion, see MICHAEL LIVINGSTON & DAVID GAMAGE, TAXATION: LAW, PLANNING, AND POLICY

92-93 (2nd ed., LexisNexis 2010). 18 Alternative reasons why employers might provide fringe benefits in lieu of cash wages might arise from regulatory pressures or employer paternalism. Id. But these motives are less important for the purposes of this Essay.



6

lower cost as compared to what individual employees could purchase on their own. And primarily as a result of the tax exclusions, employer-provided health insurance was subsidized as compared to the alternative health insurance options that employees might purchase from the market using their after-tax wages.

A) The Non-Tax Advantages of Employer-Provided Health Insurance Prior to the ACA

Insurance differs from ordinary market goods. The amount an individual pays for insurance premiums over her lifetime may be either more or less than the cost of the medical care paid for by her insurance plan, partially depending on whether the individual’s medical costs end up being higher or lower than the average medical costs incurred by similar individuals. A primary reason why individuals purchase health insurance is to protect themselves against the risk that their future medical costs may end up being higher than expected (and higher than the individual will be able to afford).

This feature of health insurance leads to the twin problems of adverse selection and risk classification. Adverse selection occurs when individuals have better knowledge about their expected future health costs than do insurance companies. Because insurance companies can only price insurance based on the insurance companies’ expectations about an individual’s future health costs, individuals who know that their future health costs are likely to be higher than insurance companies anticipate will often find that health insurance offers them a good deal. Conversely, individuals who know that their future health care costs are likely to be lower than insurance companies anticipate will often find that health insurance offers them a poor deal.19

Adverse selection results when the former (high-cost) individuals purchase more health insurance because they realize that it offers them a good deal, while the latter (low-cost) individuals purchase less health insurance because they realize that insurance offers them a relatively poor deal.20 When high-cost individuals enter an insurance market, and low-cost individuals exit the market, the inevitable result is higher costs and rising insurance premiums.21 As insurance premiums rise to reflect the higher costs of the insured pool, an ever larger group of individuals will find that insurance offers them a relatively poor deal.22 These dynamics can create “adverse selection death spirals” through repeated cycles of relatively low-cost individuals 19 Of course, whether insurance offers a “good deal” involves more than just a comparison of premiums paid to health expenses reimbursed for. Health insurance is meant to be insurance, after all, and being insured against risks can be valuable even if those risks do not end up materializing. But the comparison of premiums paid to expected reimbursements is still an important component of whether an insurance policy provides sufficient value to be worth the cost. 20 Monahan, Complex Relationship, supra note __, at 8-9. 21 E.g., Michael Rothschild & Joseph Stiglitz, Equilibrium in Competitive Insurance Markets: An Essay on the Economics of Imperfect Information, 90 Q. J. ECON 629 (1976); George A. Akerlof, The Market for "Lemons": Quality, Uncertainly and the Market Mechanism, 84 Q.J. ECON. 488, 489-90 (1970). 22 For a more in depth elaboration of an analogous dynamic, see David Gamage & Allon Kedem, Commodification and Contract Formation, Placing the Consideration Doctrine on Stronger Foundations, 73 U. CHI. L. REV. 1299, 1338-47 (2006).



7

leaving the market, leading to higher premiums, which then leads to a new group of relatively low-cost individuals leaving the market, which leads to even higher premiums, and so on.23

Highly related to adverse selection, the nature of health insurance also incentivizes insurance issuers to engage in the practice of risk classification.24 Insurance companies can use risk classification techniques to defend against adverse selection. The better information an insurance company can obtain about individuals’ future health costs, the more accurately the insurance company can price its policies to reflect the expected health costs of insured individuals.25 More problematically, if insurance companies can exclude relatively high-cost individuals from their policies, then the insurance companies can keep premiums lower while generating higher profits.26 Consequently, insurance companies invest considerable resources toward distinguishing high-cost individuals from low-cost individuals and in developing techniques for making their policies more attractive to low-cost individuals and less attractive to high-cost individuals.27 These practices can generate high administrative costs.28

Due to the problems of adverse selection and risk classification, prior to the ACA, insurance policies offered on the individual market were generally of comparatively lower quality and higher price as compared to employer-provided insurance policies.29 The “individual market” refers to when individuals purchase insurance policies directly, rather than through their employers of through government programs.30 According to one study, in 2005, “nearly 3 in 5 adults who applied for coverage in the individual market failed to find a plan they could afford because they were denied coverage, charged higher prices, or had a health problem excluded from coverage.”31 Notably, insurance companies could deny coverage to individuals with pre-existing conditions – a practice that will be banned by the ACA.32 Individuals who insurance

23 E.g., David M. Cutler & Richard J. Zeckhauser, Adverse Selection in Health Insurance, 1 FRONTIERS IN HEALTH POLICY RESEARCH (1998), available at http://www.nber.org/papers/w6107. 24 For a discussion of risk classification, see, e.g., Amy Monahan & Daniel Schwarcz, Will Employers Undermine Health Care Reform by Dumping Sick Employees?, 97 VIRGINIA L. REV. 125, Part I.A (2011). 25 David Hyman & Mark Hall, Two Cheers for Employment-Based Health Insurance, 2 YALE J. HEALTH POL’Y, L &

ETHICS 23, 31-33 (2001). 26 In my view, this aspect of risk classification presents a far more important problem than does adverse selection. I discuss adverse selection first for ease of exposition. 27 Allison Hoffman, Oil and Water: Mixing Individual Mandates, Fragmented Markets, and Health Reform, 36 AM. J. L. AND MED. 7, 28-29 (2010). 28 Id (comparing estimates for administrative costs on the individual market where risk classification is possible to administrative costs for employer-provided health insurance where risk classification is less likely); Monahan, Complex Relationship, supra note __, at n.7. 29 E.g, Michelle M. Doty et al., Failure to Protect: Why the Individual Insurance Market Is Not a Viable Option for Most U.S. Families, COMMONWEALTH FUND PUB. 1300, Feb. 2009, at 1-3; Melinda Beeuwkes Buntin et al., The Role of the Individual Health Insurance Market and Prospects for Change, 23 HEALTH AFF. 79, 81 (2004). 30 Id. 31 Hoffman, supra note __, at 53 (citing SARA R. COLLINS ET AL., THE COMMONWEALTH FUND, SQUEEZED: WHY RISING EXPOSURE TO HEALTH CARE COSTS THREATENS THE HEALTH AND FINANCIAL WELL-BEING OF AMERICAN FAMILIES 4 (2006)). 32 Monahan, Complex Relationship, supra note __, at 17.



8

companies assumed to be high risk or who had pre-existing conditions thus found it exceedingly difficult to purchase meaningful insurance on the individual market except at exorbitant costs.33

Exacerbating these problems, health insurance policies can be very complex, and individuals often find it difficult to understand exactly what coverage a policy offers with respect to the wide variety of possible future health costs the individual may incur.34 Hence, absent regulation, insurance companies often find it profitable to design policies that appeal to younger, healthier insureds while reducing benefits to insureds who develop expensive health conditions.35 And even when faced with regulations designed to prevent these practices, insurance companies have proven their ability to engage in risk classification through more subtle (and often more administratively costly) means.36

For these reasons, prior to the ACA, most commentators agreed that the individual market for health insurance was a disaster.37 In contrast, employer-provided health insurance largely solved the problems of the individual market.38 An employer’s workforce is grouped together for reasons other than their health risks, and employers can thus either market their workforce to insurance companies as a group or else self-insure their workforce with much less

33 Hoffman, supra note __, at 53. 34 E.g., Jeffrey Liebman and Richard Zeckhauser, Simple Humans, Complex Insurance, Subtle Subsidies, TAX

POLICY CENTER, February 24, 2008, available at http://www.taxpolicycenter.org/tpccontent/healthconference_zeckhauser.pdf; John Goodman, Complex Systems – Part I and Part II, NATIONAL CENTER FOR POLICY ANALYSIS: JOHN GOODMAN’S HEALTH POLICY BLOG, October 19, 2011 and November 9, 2011 respectively, available at http://healthblog.ncpa.org/complex-systems-part-i/ and http://healthblog.ncpa.org/complex-systems-part-ii/. 35 E.g., Nicholas Bagley and Jill R. Horwitz, Commentary, Why It’s Called the Affordable Care Act, 110 Mich. Law Review. First Impressions 1, 5 (2011), http://www.michiganlawreview.org/assets/fi/110/bagleyhorwitz.pdf; Peter Harbage, Too Sick for Health Care: How Insurers Limit and Deny Care in the Individual Health Insurance Market, July 20, 2009, CENTER FOR AMERICAN PROGRESS, available at http://www.americanprogress.org/issues/2009/07/too_sick.html; Beth C. Fuchs, Expanding the individual health Insurance market: Lessons from the state reforms of the 1990s, THE ROBERT WOOD JOHNSON FOUNDATION, at 10, June 2004, available at http://www.rwjf.org/files/research/20114.expandinginsurance.report.pdf. 36 E.g, Nicholas Bagley and Jill R. Horwitz, Commentary, Why It’s Called the Affordable Care Act, 110 Mich. Law Review. First Impressions 1, 6 (2011), http://www.michiganlawreview.org/assets/fi/110/bagleyhorwitz.pdf; THE

AMERICAN ACADEMY OF ACTUARIES, Risk Classification in the Voluntary Individual Health Insurance Market, March 2009, available at http://www.actuary.org/pdf/health/risk_mar09.pdf; Thomas F. Wildsmith, et al., THE

AMERICAN ACADEMY OF ACTUARIES, Risk Classification in Individually Purchased Voluntary Medical Expense Insurance, February 1999, available at http://actuary.org/pdf/health/risk.pdf . 37 E.g.,Michelly Doty, et al., Failure to Protect: Why the Individual Insurance Market Is Not a Viable Option for Most U.S. Families, THE COMMONWEALTH FUND, July 21, 2009, Volume 62, available at http://www.commonwealthfund.org/Publications/Issue-Briefs/2009/Jul/Failure-to-Protect.aspx; Bethanne Fox, New Report: Individual Health Insurance Market Failing Consumers, THE COMMONWEALTH FUND, July 21, 2009, available at http://www.commonwealthfund.org/News/News-Releases/2009/Jul/New-Report-Individual-Health-Insurance-Market-Failing-Consumers.aspx. 38 David Hyman & Mark Hall, Two Cheers for Employment-Based Health Insurance, 2 YALE J. HEALTH POL’Y, L &

ETHICS 23, 32-35 (2001).



9

risk of adverse selection.39 Employer-provided coverage similarly minimizes the costs of risk classification. Although employers could in theory exclude their high-cost employees from insurance policies or charge those employees more, the rest of the employee’s workforce might learn about these practices, which could harm employee morale and retention. In practice, employers appear to mostly provide health care policies that benefit their entire workforce.40

Moreover, employers can and do assist their employees with the complexity of the health insurance decision-making process by functioning as intermediaries.41 Employers’ human resources departments support employees by screening health insurance options, helping employees select from among the options provided, and aiding employees who have disputes about reimbursements or coverage provided.42

In sum, prior to the ACA, employers could offer their employees better health insurance options at considerably lower costs than what was available on the individual market. Whereas the individual market suffered from adverse selection, risk classification, and other information problems, employers could largely solve these problems by creating insurance groups unrelated to health costs and by functioning as information intermediaries.43 Thus, a major reason why employers offered their employees subsidized health insurance was that employers’ advantages in providing health insurance made that insurance more valuable to employees than the foregone cash wages.

B) The Tax Advantages of Employer-Provided Health Insurance Prior to the ACA

Not only was employer-provided health insurance generally cheaper and better than what could be purchased on the individual market prior to the ACA, but employer-provided health insurance was also significantly tax advantaged as compared to insurance purchased on the individual market. The primary sources of this tax advantage were the tax exclusions for employer-provided health insurance.44 Employees who received subsidized health insurance

39 Stuart M. Butler, Evolving Beyond Traditional Employer-Sponsored Health Insurance, HAMILTON PROJECT

DISCUSSION PAPER 2007-06, at 7 (May 2007). 40 Id., David Hyman & Mark Hall, Two Cheers for Employment-Based Health Insurance, 2 YALE J. HEALTH POL’Y, L & ETHICS 23, 30 (2001). (“Surveys and focus groups indicate that employers do a reasonably good job reflecting their workers' values and preferences, just as one would expect in a reasonably competitive labor market.”). 41 David Hyman & Mark Hall, Two Cheers for Employment-Based Health Insurance, 2 YALE J. HEALTH POL’Y, L &

ETHICS 23, 33-35 (2001). 42 Id. 43 But see Stuart M. Butler, Evolving Beyond Traditional Employer-Sponsored Health Insurance, HAMILTON

PROJECT DISCUSSION PAPER 2007-06, at 8-10 (May 2007) (arguing that small employers do not enjoy the same advantages with respect to providing insurance as do large employers and that relying on employer-provided insurance creates other problems such as interfering with job mobility). 44 E.g., Stan Dorn, Capping the Tax Exclusion of Employer-Sponsored Health Insurance: Is Equity Feasible?, THE

URBAN INSTITUTE, June 2009, available at http://www.urban.org/uploadedpdf/411894_cappingthetaxexclusion.pdf.



10

from their employers could exclude the value of those subsidies from taxable income.45 Employer-subsidized health insurance was also excludable from payroll taxes.46 Moreover, by having their employers establish a cafeteria plan under section 125 of the Internal Revenue Code, employees could also reduce their taxable income by the amounts the employees contributed to pay for health insurance premiums.47 Hence, even in the absence of any employer subsidies, employees could pay for employer-provided health insurance entirely with pre-tax dollars.48

The exclusion for employer-provided health insurance was the largest federal tax expenditure in 2007.49 The Joint Committee on Taxation estimated the value of the exclusion to be worth $246.1 billion annually.50

Self-employed individuals could also enjoy similar healthcare tax benefits to those available for employees.51 Most importantly, self-employed individuals could deduct health insurance payments for them and for their dependents under section 162(l) of the Internal Revenue Code.52 The rules for self-employed individuals mostly mirrored the rules for individuals receiving employer-provided health insurance.53 However, a taxpayer could not claim the self-employed health insurance deduction for any month in which either the taxpayer or the taxpayer’s spouse was eligible to participate in an employer-sponsored health plan.54 The

45 I.R.C. § 106(a); I.R.C. § 105(b). For a broader discussion of the exclusion, see MICHAEL LIVINGSTON & DAVID

GAMAGE, TAXATION: LAW, PLANNING, AND POLICY 98-99 (2nd ed., LexisNexis 2010); Peter J. Wiedenbeck, Taxes and Healthcare, 124 TAX NOTES 889, 889-90 (2009). 46 Wiedenbeck, supra note __, at 892. 47 JOINT COMMITTEE ON TAXATION, TAX EXPENDITURES FOR HEALTH CARE, JCX-66-08, at 6, (July 30, 2008), available at http://www.jct.gov/publications.html?func=startdown&id=1193. 48 Monahan, Complex Relationship, supra note __, at 3. 49 JOINT COMMITTEE ON TAXATION, TAX EXPENDITURES FOR HEALTH CARE, JCX-66-08, at 2, (July 30, 2008), available at http://www.jct.gov/publications.html?func=startdown&id=1193.

For discussions of the “tax expenditure” concept, see, e.g., Edward Kleinbard, The Congress Within the Congress: How Tax Expenditures Distort our Budget and Our Political Processes, 36 OHIO NORTHERN L. REV. 1, 3

(2010); Daniel Shaviro, Rethinking Tax Expenditures and Fiscal Language, 57 TAX L. REV. 187, 187 (2004). 50 JOINT COMMITTEE ON TAXATION, TAX EXPENDITURES FOR HEALTH CARE, JCX-66-08, at 2, (July 30, 2008), available at http://www.jct.gov/publications.html?func=startdown&id=1193. 51 Supplementing the exclusion for employer-provided health insurance and the self-employed health insurance deduction were a number of other tax benefits related to employer-provided healthcare. However, as these benefits are less important for the purposes of this Essay, I will not discuss them here. For discussions of these other tax benefits, see e.g., Fred Goldberg & Susannah Camic, Legal Solutions in Health Reform: Tax Credits for Health Insurance, at 4-6 (2009), available at www.oneillinstitute.org/projects/reform/Tax_Credits.html; JOINT COMMITTEE

ON TAXATION, TAX EXPENDITURES FOR HEALTH CARE, JCX-66-08 (July 30, 2008), available at http://www.jct.gov/publications.html?func=startdown&id=1193; Wiedenbeck, supra note __. 52 Fred Goldberg & Susannah Camic, Legal Solutions in Health Reform: Tax Credits for Health Insurance, at 4-5 (2009), available at www.oneillinstitute.org/projects/reform/Tax_Credits.html. 53 Id. at 5. 54 JOINT COMMITTEE ON TAXATION, TAX EXPENDITURES FOR HEALTH CARE, JCX-66-08, at 18, (July 30, 2008), available at http://www.jct.gov/publications.html?func=startdown&id=1193.



11

Joint Committee on Taxation estimated the value of the health insurance deduction for self-employed individuals to be worth $4.8 billion in 2007.55

In contrast, no equivalent tax benefits were available for unemployed taxpayers or for employed taxpayers whose employers did not offer health insurance.56 There was no generally applicable tax deduction or exclusion for healthcare expenditures by unemployed taxpayers or by employed taxpayers lacking employer-provided health insurance.57 These taxpayers could claim an itemized deduction to the extent that their unreimbursed medical expenses exceeded 7.5 percent of adjusted gross income.58 But this deduction required taxpayers to forgo the standard deduction, and the 7.5 percent threshold made the deduction of minimal value for most taxpayers.59 The Joint Committee on Taxation estimated the value of this limited itemized deduction for medical expenses to be only $8.7 billion in 2007, as compared to the combined value of $250.9 Billion for the exclusion for employer-provided health insurance and the self-employed health insurance deduction.60

The net result was a significant tax advantage for employer-provided health insurance as compared to insurance purchased on the individual market. Consider an example of this tax disparity (quoted from Amy Monahan):

“The tax benefit for [employer-sponsored] coverage provides a significant subsidy for most taxpayers. For example, assume that Taxpayer A and Taxpayer B desire the same insurance coverage, an individual policy that costs $3,750. Taxpayer A is offered her desired coverage through her employer, while Taxpayer B is not. Both taxpayers are in the 25% marginal rate bracket. Taxpayer A needs to earn only $3,750 in wages to purchase such coverage, while Taxpayer B must earn $5,000 in wages to have sufficient after-tax funds available for his purchase. If we take into account payroll taxes of 7.65% and an assumed state income tax rate of 5%, the amount of wages necessary to pay for a $3,750 policy rises to $5,162. Taxpayer A receives an effective subsidy of $1,412 to purchase her health insurance coverage, solely because her employer makes such coverage available to her, and regardless of whether her employer makes any contribution toward such coverage.”61

55 Id. 56 However, taxpayers whose incomes were sufficiently low might qualify for Medicaid. 57 Monahan, Complex Relationship, supra note __, at 3; JOINT COMMITTEE ON TAXATION, TAX EXPENDITURES FOR

HEALTH CARE, JCX-66-08, at 13, (July 30, 2008), available at http://www.jct.gov/publications.html?func=startdown&id=1193. 58 I.R.C. Section 213. The 7.5% threshold for deducting medical expenses will increase to 10% in 2013, pursuant to Section 9013(d) of the Patient Protection and Affordable Care Act, amending I.R.C. § 213(a). However, through 2016, taxpayers of age 65 and older will continue to be able to use the old 7.5% threshold, pursuant to Section 9013(d) of the Patient Protection and Affordable Care Act, amending I.R.C. § 213(f). 59 JOINT COMMITTEE ON TAXATION, TAX EXPENDITURES FOR HEALTH CARE, JCX-66-08, at 22-24, (July 30, 2008), available at http://www.jct.gov/publications.html?func=startdown&id=1193. 60 Id. at 2. 61 Monahan, Complex Relationship, supra note __, at 3-4.



12

II) HOW THE ACA WILL AFFECT INCENTIVES TO OFFER HEALTH INSURANCE

Prior to the ACA, employers offered health insurance to their individual employees both because employers could provide better quality health insurance at lower cost than what was available on the individual market and because employer-provided health insurance was significantly tax advantaged. As this Part explains, the ACA will dramatically alter both of these incentives. Once key provisions of the ACA go into effect in 2014, employers will face a very different calculus in deciding whether to offer health insurance to their individual employees.

A) The ACA’s Improvements to the Health Insurance Available on the Individual Market

A primary goal of the ACA was to fix the problems plaguing the individual market for health insurance.62 Most critically, the ACA will impose open enrollment63 and guaranteed renewal64 requirements on all health insurance plans offered in the individual and small group markets so that these plans must accept all applicants for health insurance.65 The ACA will further limit insurance issuers’ ability to charge applicants different prices based on their expected health risks. The ACA will only allow health insurance issuers to vary their prices based on four factors: the size of the applicant’s family (for applicants seeking family coverage), the geographic region in which the applicant resides, the applicant’s age, and whether the applicant uses tobacco.66 Even with respect to these factors, insurance issuers will be limited to charging their oldest applicants no more than three times the prices charged to their youngest applicants and to charging tobacco users no more than one and a half times the prices charged to non-smokers.67 In effect, the ACA will prevent insurance plans from discriminating against applicants with pre-existing health conditions.

More generally, these provisions of the ACA will significantly limit insurers’ ability to engage in risk classification. Yet, without further regulation, limiting insurers’ ability to engage in risk classification could exacerbate problems related to adverse selection, potentially

62 See Tom Baker, Health Insurance, Risk, and Responsibility After the Patient Protection and Affordable Care Act, U OF PENN, INST FOR LAW & ECON RESEARCH PAPER NO. 11-03, at 7-12 (2011) (forthcoming, 159 U. PENN. L. REV. 1577), available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1759366. 63 PPACA § 1201 (2010) (amending the Public Health Service Act to include § 2702, requiring that every health insurance issuer accept all applicants.) 64 PPACA § 1201 (2010) (amending the Public Health Service Act to include § 2703(a), “Except as provided in this section, if a health insurance issuer offers health insurance coverage in the individual or group market, the issuer must renew or continue in force such coverage at the option of the plan sponsor or the individual, as applicable.”) 65 Baker, supra note __, at 10. 66 PPACA § 1201 (2010) (amending the Public Health Service Act to forbid price discrimination except on the basis of age, family status, rating area, or tobacco use.) 67 PPACA § 1201 (2010) (amending the Public Health Service Act to include § 2701, specifying that discrimination based on age “not vary by more than 3 to 1 for adults (consistent with section 2707(c))” and that discrimination based on tobacco use “not vary by more than 1.5 to 1.”)



13

undermining the entire individual market. The ACA thus also includes a number of provisions meant to combat adverse selection. The most important of these is the individual mandate created by new Section 5000A of the Internal Revenue Code.68 The individual mandate establishes penalties for citizens who do not obtain health insurance constituting “minimum essential coverage.”69 In essence, the individual mandate is intended to incentivize healthier individuals to obtain insurance coverage so as to prevent adverse selection problems.

In order to further facilitate reforming the individual market for health insurance, the ACA directs the states to create and administer Affordable Insurance Exchanges (“Exchanges”).70 If a state does not create an Exchange, then the Department of Health and Human Services (“HHS”) will be empowered to establish and operate an Exchange on behalf of the State.71 The ACA instructs the states to establish risk adjustment mechanisms so that insurance plans that end up with a disproportionately low-risk group of insureds will be assessed charges to be used to compensate plans that end up with a disproportionately high-risk group of insureds.72 The Exchanges will likely administer these risk adjustment policies.73 Additionally, the Exchanges will regulate the insurance policies offered to consumers and will act as information intermediaries to aid consumers in selecting insurance plans.74

The ACA makes generous premium tax credits available for low and moderate income taxpayers purchasing health insurance from an Exchange.75 When combined with the individual mandate, the premium tax credits should help to ensure that there will be a critical mass of insureds purchasing health insurance from the Exchanges. The premium tax credits also serve an important role in making health insurance affordable, which is particularly important since the individual mandate will require individuals to purchase health insurance.76 Whereas the individual mandate combats adverse selection by penalizing taxpayers who do not purchase

68 PPACA § 1501(b) (2010) (amending the Internal Revenue Code of 1986 by adding Sec. 5000A”). For discussions of the individual mandate, see Hoffman, supra note __; Edward A. Zelinsky, The Health-Related Tax Provisions of PPACA and HCERA: Contingent, Complex, Incremental and Lacking Cost Controls, CARDOZO

LEGAL STUDIES RESEARCH PAPER NO. 301, at 17-26, (2010), available at http://ssrn.com/abstract=1633556. 69 Baker, supra note __, at 8. 70 PPACA § 1311(2010); Prop. Treas. Reg. § 36B, 76 Fed. Reg. 50931, 50932 (Aug. 17, 2011). 71 PPACA § 1321(2010). 72 PPACA § 1343 (2010) (instructing the states to create risk adjustment mechanism for insurance plans in the individual and small group market). 73 Baker, supra note __, at 12. 74 Id. 75 I.R.C. § 36B. See also Peter Gosselin, New Rule Could Narrow Aid for Heal-Plan Buyers and Shrink Insurers’ Sales, BLOOMBERG GOVERNMENT STUDY 3 (September 27, 2011) (“The premium tax credits are the chief means by which the overhaul law will help middle-class Americans − those with household incomes between $22,350 and $89,400 for a family of four this year – afford health coverage.). 76 The individual mandate has an affordability exemption – I.R.C. § 5000A(e)(1). But “affordability” is determined after accounting for the subsidy provided by the premium tax credits – I.R.C. § 5000A(e)(1)(B)(ii). The premium tax credits thus plan an important role in supporting the policy goals of the individual mandate because without the premium tax credits many low and moderate income taxpayers would be exempt from the mandate because they would not have “affordable” insurance options.



14

health insurance (i.e., sticks), the premium tax credits incentivize taxpayers to purchase health insurance (i.e., carrots).

To the extent these policies prove effective, the Exchanges should at least partially solve the problems that previously afflicted the individual market for health insurance. Prior to the ACA, employer-provided health insurance was superior to individual market offerings because employers were able to mitigate risk classification and adverse selection problems and to act as information intermediaries. Similarly, the market reform provisions of the ACA are designed to combat risk classification problems, while the individual mandate and premium tax credits are designed to mitigate adverse selection problems. The Exchanges should further mitigate both risk classification and adverse selection problems while functioning as information intermediaries.

It remains to be seen how well these provisions of the ACA will function in fixing the problems that previously plagued the individual market. If these provisions are sufficiently effective, insurance policies offered on the Exchanges could potentially turn out to be of better quality and lower cost than employer-provided offerings. Perhaps more likely, if the provisions are only partially effective, employer-provided insurance might retain its advantages over insurance policies offered on the Exchanges, but with the advantages of employer-provided insurance being significantly reduced as compared to the advantages employer-provided insurance previously enjoyed over the insurance policies available on the individual market prior to the ACA.

B) The ACA’s Modifications to the Tax Exclusions and Creation of Exchange Subsidies

The ACA will mostly leave the tax benefits for employer-provided health insurance intact.77 Most importantly, the ACA retains both the tax exclusion for employer-provided health insurance and the self-employed health insurance deduction.78 The primary way in which the ACA will alter the tax favorability of employer-provided health insurance is through the creation of premium tax credits to subsidize the purchase of insurance policies from the Exchanges.

Before assessing the impact of the premium tax credits, though, it is worth briefly discussing some of the changes the ACA will make to the previously existing tax benefits for employer-provided health insurance. To address criticisms that the exclusion for employer-

77 E.g., Linda Blumberg, et al., Why Employers Will Continue to Provide Health Insurance: The Impact of the Affordable Care Act, THE URBAN INSTITUTE, October 2011, available at http://www.urban.org/UploadedPDF/412428-The-Impact-of-the-Affordable-Care-Act.pdf; The HR Specialist, Health care reform: Will employers keep offering coverage?, BUSINESS MANAGEMENT DAILY, May 10, 2011, available at http://www.businessmanagementdaily.com/14501/health-care-reform-will-employers-keep-offering-coverage. 78 Zelinsky, supra note __, at 6.



15

provided health insurance leads to overconsumption of health care services,79 the ACA includes an excise tax on “Cadillac” health plans.80 The “Cadillac” tax will impose a forty percent levy on employer-provided insurance policies that benefit from the tax exclusion and that provide “excess benefit” – with “excess benefit” defined as an outcome where the annual cost of a health insurance plan exceeds a specified threshold designed to be higher than the amount most individuals and families pay for health insurance.81 The excise tax on “Cadillac” health plans thus reduces the tax benefit that the exclusions for employer-provided health insurance generate for high-cost health plans.

However, the “Cadillac” excise tax is not scheduled to go into effect until 2018,82 and there is reason to doubt whether Congress and the President83 will allow the provision to go into effect at that time.84 In addition, even if and when the “Cadillac” excise tax becomes active, the tax will not affect employer-provided health insurance plans unless the plans both benefit from the tax exclusion and have costs that exceed the excess-benefit threshold.85 Consequently, the exclusion for employer-provided health insurance will continue to generate a large tax-subsidy even after 2018.86

It is also noteworthy that, beginning in 2013, the ACA will raise the itemized medical expenses deduction percentage threshold from 7.5% to 10%.87 Otherwise, the changes that the ACA will make to the previously existing health care tax benefits are not of primary importance for the purposes of this Essay.88

What will be of primary importance are the premium tax credits to be provided by new Section 36B of the Internal Revenue Code starting in 2014.89 The premium tax credits will defray the cost of health insurance for qualifying low and middle income taxpayers purchasing

79 Id. at 5 (“This tax originated in the broad consensus that the Code’s current tax treatment of employer-provided medical care encourages overconsumption of medical services.”) 80 I.R.C. § 4980I, added by PPACA §§ 9001 and 10901 and HCERA § 1401. 81 Zelinsky, supra note __, at 6-7. 82 HCERA § 1401(b). 83 At the time of this writing, it remains to be seen who will be the President in the years leading up to 2018. 84 See, e.g., Zelinsky, supra note __, at 8 (“Given the palpable reluctance of President Obama and the members of the 111th Congress to force their constituents to confront the tax on ‘Cadillac’ plans any time soon, why should we expect a future President and the senators and representatives of the 115th Congress to let this tax go into effect in 2018?”). 85 I.R.C. § 4980I(b)(3)(C). 86 E.g., Shubham Singhal, et al., How US health care reform will affect employee benefits, MCKINSEY QUARTERLY –

MCKINSEY & COMPANY, June 2011, available at www.mckinseyquarterly.com/How_US_health_care_reform_will_affect_employee_benefits_2813. 87 I.R.C. § 213(a), as amended by PPACA § 9013. However, through 2016, a taxpayer of age of 65 or older can use the prior threshold of 7.5%. 88 For instance, the ACA makes a number of minor changes to the rules governing HSAs, MSAs, HRAs, and FSAs. Zelinsky, supra note __, at 10-11. 89 I.R.C. § 36B, added by PPACA §§ 1401(a) and 10105 and HCERA §§ 1001 and 1004.



16

health insurance from an Exchange.90 The amounts by which the premium tax credits will defray the costs of health insurance will depend on: a taxpayer’s household income as compared to the federal poverty line, the cost of health insurance premiums for the applicable benchmark plan, and the number of eligible members in the taxpayer’s coverage family.91 As a baseline, taxpayers whose household incomes are less than 133 percent of the federal poverty line will be expected to contribute no more than 2 percent of their household incomes toward health insurance;92 taxpayers whose household incomes are between 300 and 400 percent of the federal poverty line will be expected to contribute no more than 9.5 percent of their household incomes toward health insurance; and taxpayers whose household incomes are between these levels will be expected to contribute maximum amounts of between 2 and 9.5 percent of their household incomes, on a sliding scale.93 Taxpayers whose household incomes are above 400 percent of the federal poverty line will not be eligible for the premium tax credits.94

As a supplement to the premium tax credits, taxpayers purchasing insurance from an Exchange may also be eligible for cost-sharing subsidies.95 The cost-sharing subsidies will reduce taxpayers’ out-of-pocket costs for deductibles, coinsurance, copayments, and similar amounts that would otherwise be charged to them by their health insurance plans. Like the premium tax credits, the value of the cost sharing subsidies a taxpayer will be eligible for depends on the taxpayer’s household income as a percent of the federal poverty line.96

To understand the value provided by the premium tax credits and cost sharing subsidies, the following chart – adapted from a study by the Tax Policy Center – estimates the total federal subsidies a family of four would be eligible for in 2016 for health insurance purchased from an Exchange:97

90 More specifically, taxpayers must have household income for the taxable year of between 100 percent and 400 percent of the federal poverty line for the taxpayer’s family size in order to be eligible for the premium tax credits. Prop. Treas. Reg. § 36B, 76 Fed. Reg. 50931, 50933 (Aug. 17, 2011). 91 For a more detailed description of credit computation, see Prop. Treas. Reg. § 36B, 76 Fed. Reg. 50931, 50933-34 (Aug. 17, 2011). 92 However, note that most taxpayers whose household incomes are below 133% of the federal poverty line should be eligible for Medicaid and not the premium tax credits. 93 I.R.C. § 36B(b)(3)(A)(i). These percentage amounts may be indexed for inflation or excess premium growth pursuant to I.R.C. § 36B(b)(3)(A)(ii). 94 I.R.C. § 36B(c)(1)(A). 95 PPACA § 1402. 96 Id. 97 Stephanie Rennane & C. Eugene Steurle, Healh Reform: A Two-Subsidy System, Urban Institute and Brookings Institution: Tax Policy Center S10-001, Table 3 (2010), available at http://www.taxpolicycenter.org/library/displayatab.cfm?Docid=2699.



17

As the chart demonstrates, low- and moderate-income families who purchase health insurance from an Exchange may be eligible for subsidies worth many thousands of dollars annually. However, the premium tax credits will not be available to taxpayers who have the option of purchasing “affordable” employer-sponsored health insurance.98 An employer who offers employees the option of purchasing “affordable” health insurance may thus make those employees ineligible for the premium tax credits, regardless of whether the employees actually enroll in the employer-sponsored health insurance.99 An offer of employer-sponsored health insurance will be considered “affordable” for purposes of a taxpayer’s eligibility for the premium

98 I.R.C. § 36B(c)(2)(B)-(C); Prop. Treas. Reg. § 36B, 76 Fed. Reg. 50931, 50935 (Aug. 17, 2011).

Taxpayers eligible for health insurance from other government programs will also generally be ineligible for the premium tax credits. I.R.C. § 36B(c)(2)(B); Prop. Treas. Reg. § 36B, 76 Fed. Reg. 50931, 50934-35 (Aug. 17, 2011). 99 For an employer’s offer of health insurance to disqualify employees from receiving the premium tax credits, the offered health insurance must be both “affordable” (id.) and provide “minimum value” – I.R.C. § 36B(c)(2)(C)(ii). A full discussion of the minimum value rule is beyond the scope of this essay; but it is worth noting that the purpose of the minimum value rule is to insure that employer-provided insurance must have some real content in order to protect the employer from the employer-mandate penalties of I.R.C. § 4980H and in order to disqualify employees from receiving the premium tax credits.

Household Income as a % of FPL

Household Income: Cash Compensation

Amounts

Value of Premium Tax

Credits Received

Value of Cost Sharing

Subsidies Received

Total Value of Health Insurance Subsidies

100 24,000$ 13,598$ 4,834$ 18,433$ 125 30,000$ 13,473$ 4,834$ 18,307$ 150 36,000$ 12,595$ 3,021$ 15,617$ 175 42,000$ 11,738$ 3,021$ 14,759$ 200 48,000$ 10,940$ 604$ 11,544$ 225 54,000$ 9,869$ 604$ 10,473$ 250 60,000$ 9,053$ -$ 9,053$ 275 66,000$ 7,776$ -$ 7,776$ 300 72,000$ 6,952$ -$ 6,952$ 325 78,000$ 6,468$ -$ 6,468$ 350 84,000$ 5,761$ -$ 5,761$ 375 90,000$ 5,165$ -$ 5,165$ 400 96,000$ 4,570$ -$ 4,570$ 425 102,000$ -$ -$ -$ 450 108,000$ -$ -$ -$

Estimated Value of Exchange Health Insurance Subsidies(Family of Four, 2016)



18

tax credits if the amount the taxpayer would have to contribute to pay for the insurance premiums exceeds 9.5 percent of the taxpayer’s household income.100

Crucially, an offer of “affordable” employer-sponsored health insurance will result in an employee’s entire family being ineligible for the premium tax credits, not just the employee.101 Moreover, under the Treasury Department’s proposed regulations, whether an employer’s offer of family coverage is considered “affordable” is determined based on the cost the employee would need to contribute for self-only coverage.102 In other words, if an employer offers an insurance policy with an option for family coverage, and if the amount an employee would need to contribute to pay for the portion of the policy covering only the employee (and not also the other members of the employee’s family) is less than 9.5 percent of the employee’s household income, then the employee’s entire family will be ineligible for the premium tax credits.

The policy rationale for basing the “affordability” of family coverage on the cost of self-only coverage for the employee is (of course) the goal of federal budgetary savings.103 I will elaborate on this policy decision further in Part III.A of this Essay.104 For now, it is important to understand that an employer who offers an option to purchase family coverage, and who offers self-only coverage costing less than 9.5 percent of an employee’s household income, thereby makes the employee’s family members ineligible for the premium tax credits – regardless of the cost of the family coverage.

In effect, then, the ACA makes employers choose between offering their employees either the benefit of the tax exclusions or the premium tax credits. An employee must enroll in employer-sponsored health insurance in order to receive the tax exclusions, but enrolling in employer-sponsored health insurance makes the employee ineligible for the premium tax credits.105 Moreover, employers cannot leave the choice of whether to take advantage of the tax exclusions or the premium tax credits to their individual employees, as even offering an employee the option of “affordable” employer-provided health insurance makes the employee ineligible for the premium tax credits.106

This dynamic is significant because whereas the premium tax credits and cost sharing subsidies are more valuable for lower income taxpayers, the exclusions from income taxes and

100 I.R.C. § 36B(c)(2)(C)(i)(II); Prop. Treas. Reg. § 36B, 76 Fed. Reg. 50931, 50935 (Aug. 17, 2011). 101 Id. 102 Prop. Treas. Reg. § 36B, 76 Fed. Reg. 50931, 50935 (Aug. 17, 2011). 103 See, e.g., Sarah Kliff, Health Reform’s $50 Billion Question: What’s ‘Affordable’?, Blog Post, EZRA KLEIN: ECONOMIC AND DOMESTIC POLICY, AND LOTS OF IT, August 16, 2011 (quoting Tim Jost as saying “I don’t think they want the headlines that it was going to cost $50 billion more…”). 104 Notes __ and accompanying text infra. 105 I.R.C. § 36B(c)(2)(C)(iii). 106 I.R.C. § 36B(c)(2)(B)-(C).



19

payroll taxes are more valuable for higher income taxpayers.107 Comparing just the Exchange subsidies108 to the tax exclusions109 based on the Tax Policy Center’s estimates for 2016,110 the breakeven point for an individual might be when household income is somewhere between 350 percent and 375 percent of the federal poverty line.111 And the breakeven point for a family of four might be when household income reaches 400 percent of the federal poverty line.112 For household incomes below these breakeven points, the Exchange subsidies will generally offer more value than the tax exclusions. Conversely, for household incomes above these breakeven points, the tax exclusions will generally offer more value than the Exchange subsidies.

These breakeven analyses assume that the health-insurance policies offered on the Exchanges will be of equivalent cost and quality as compared to employer-sponsored health insurance policies. If all of the available Exchange coverage options end up being inferior to employer-sponsored health insurance options, then the breakeven thresholds would need to be adjusted accordingly.113 For simplicity, in this Essay, I will mostly follow the Tax Policy Center’s approach and assume that Exchange health insurance and employer-provided health insurance will be of equivalent cost and value after the relevant provisions of the ACA come into effect in 2014.114

These breakeven analyses also assume that employers offer their employees subsidized health insurance as a form of employee compensation,115 an assumption that underlies virtually all economic studies of health care provision.116 When comparing the premiums they must pay for Exchange coverage against the premiums they must pay for employer-sponsored coverage, many employees will prefer employer-sponsored coverage to the extent that employers continue to subsidize this coverage. But in the long run, these employer subsidies come out of the 107 See Rennane & Steurle, supra note __, at Table 4 (comparing the value of the Exchange subsidies and tax exclusions for different household income levels). 108 By “Exchanges subsidies” I mean both the premium tax credits and the cost sharing subsidies. 109 By “tax exclusions” I mean both the income tax exclusion and the payroll tax exclusion for employer-provided health insurance. 110 Rennane & Steurle, supra note __, at Table 4. Note that I have subtracted the employer penalty from the Tax Policy Center’s analysis in calculating the breakeven thresholds. 111 Id. 112 Id. 113 Note that there will be a variety of insurance policies offered on the Exchanges, with different tradeoffs between premium costs and generosity of benefits. Although some of these options will undoubtedly be of inferior quality to employer-sponsored offerings, a well-run Exchange should be able to provide health insurance options competitive with those available from employers. The Exchanges are very different from previous government health care programs, like Medicaid and Medicare, in this respect. 114 Rennane & Steurle, supra note __, at Tables 1 & 3. 115 See notes ___ supra and accompanying text. 116 See, e.g., Devon Herrick, Health Exchange Subsidies Will Reduce Employer Health Plans, NATIONAL CENTER

FOR POLICY ANALYSIS BRIEF 758 (2011), available at http://www.ncpa.org/pub/ba758 (“Economists generally agree employee benefits are a dollar-for-dollar substitute for wages.”); Katherine Pratt, Funding Health Care With An Employer Mandate: Efficiency and Equity Concerns, 39 ST. LOUIS U. L.J. 155, 161-62 (1994) (“Economists agree that employees ultimately bear the economic burden of employer-provided health care benefits, in the form of lower wages.”).



20

amounts paid as wages to employees.117 As such, these employer subsidies should be factored out when evaluating the effective taxes facing employees.

Hence, were the tax exclusions and the Exchange subsidies the only relevant factors, employers would face incentives to offer “affordable” health insurance only to their higher-income employees – to those employees whose household incomes are above the relevant breakeven thresholds. Both lower-income employees and their employers would jointly benefit from the employers not offering the lower-income employees “affordable” health insurance. Employers who would have subsidized health insurance for their lower-income employees, were it not for the Exchange subsidies, could provide better value for those employees by instead using the amounts of the subsidies to increase the lower-income employees’ cash wages.

C) The ACA’s New Employer-Mandate Penalties and Nondiscrimination Rules

Beyond the tax exclusions and Exchange subsidies, the ACA contains other provisions that will significantly affect employers’ incentives as to whether to offer health insurance. In drafting the ACA, the Obama Administration and the Democratic majority in Congress were very concerned about whether employers would stop offering health insurance after the ACA came into effect. The Administration campaigned for the ACA by telling the public that anyone who liked their existing insurance coverage would be able to keep it.118 The Administration also campaigned for the ACA based on budget estimates that the ACA would reduce the deficit.119 Maintaining the previous system of employer-sponsored coverage for lower-income taxpayers was considered important for realizing the ACA’s deficit-reducing potential because additional lower-income employees qualifying for the Exchange subsidies would drive up the budgetary cost of the Exchange subsidies.

The ACA was thus drafted to include additional provisions incentivizing employers to maintain employer-sponsored health insurance even for their lower-income employees. The two most important of these provisions are the employer-mandate penalties created by new Section 117 Id. 118 E.g., David A. Hyman, Employment-Based Health Insurance: Insurance: Is Health Reform a “Game Changer?” 1 N.Y.U. REV. EMP. BENEFITS & EXECUTIVE COMPENSATION 1A-1 (2010) (“During the 2008 campaign, (then Senator) Obama routinely promised ‘if you like your coverage you can keep it.’ Even ABC News thought the promise was ‘not literally true,’ but Senator Obama had found a winning slogan, and he stuck to it. President Obama repeated and expanded this claim during the battle over health reform, flatly claiming in a speech to the AMA that, ‘no matter how we reform health care, we will keep this promise: If you like your doctor, you will be able to keep your doctor. Period. If you like your health care plan, you will be able to keep your health care plan. Period. No one will take it away. No matter what.’”). 119 E.g., David Hyman, PPACA in Theory and Practice: The Perils of Parallelism, VIRGINIA LAW REVIEW, at 18 (forthcoming) (“PPACA wouldn’t have passed unless it was deemed to be “affordable” – which in Washington meant that it had to be scored by the Congressional Budget Office (“CBO”) as costing less than $1 trillion over the ten year budgetary window. The task of selling PPACA to a skeptical public would also be substantially easier if it could somehow be scored by the CBO as deficit reducing (at least over the same budgetary window). As I have noted elsewhere, the Administration and Congress hit both of these targets by completely gaming the CBO scoring process.”); http://voices.washingtonpost.com/ezra-klein/2010/07/omb_aca_cbo_and_the_deficit.html



21

4980H of the Internal Revenue Code and the nondiscrimination rules created by new Section 2716 of the Public Health Service Act.120

The employer-mandate penalties of new Section 4980H of the Internal Revenue Code will apply to “applicable large employers” beginning in 2014.121 With a few exceptions,122 employers who employ at least 50 full-time employees for more than 120 days during a calendar year will be considered applicable large employers potentially subject to the Section 4980H penalties.123

Section 4980H includes two separate employer-mandate penalties, contained respectively in 4980H(a) and in 4980H(b). The 4980H(a) penalty applies when an applicable large employer “fails to offer its full-time employees (and their dependents) the opportunity to enroll” in qualifying employer-sponsored health coverage.124 For any month in which it is triggered, the 4980H(a) penalty will equal one-twelfth of $2,000 multiplied by the total number of full-time employees employed during the month. The 4980H(a) penalty is thus not tailored to the number of employees to whom an employer does not offer health insurance. Instead, the 4980H(a) penalty will be assessed based on an employer’s total number of full-time employees, even if the employer provides health insurance to all but a small percentage of those employees.125

The 4980H(a) penalty will thus interfere with the strategy of an employer providing health insurance only to higher-income employees and sending lower-income employees to the Exchanges. Yet the 4980H(a) penalty will not prevent a variation of this strategy, as the penalty will only be triggered if an employer fails to offer health insurance. An employer will thus be able to avoid the 4980H(a) penalty by offering health insurance to all full-time employees regardless of how much the employees would be charged for that insurance. Therefore, an 120 Also noteworthy is the small-employer health insurance tax credit provided by new Section 45R of the Internal Revenue Code. For a discussion of this tax credit, see Zelinsky, supra note __, at 14-17. 121 I.R.C. § 4980H; Mireille Khoury, INTERNAL REVENUE SERVICE, Notice 2011-73 – Request for Comments on Health Coverage Affordability Safe Harbor for Employers (Section 4980H), October 3, 2011, available at http://www.irs.gov/irb/2011-40_IRB/ar11.html. 122 E.g., I.R.C. § 4980H(c)(2)(B)(i)(II). 123 I.R.C. § 4980H(c)(2). Whereas applicable large employers are incentivized to offer health insurance through the Section 4980H penalties, small employers are incentivized to offer health insurance through tax credits offered by new Section 45R of the Internal Revenue Code. I will not discuss the Section 45R small-employer health insurance tax credits, as they are less important for the purposes of this Essay. For a discussion of these credits, see Zelinsky, supra note __, at 14-17. 124 I.R.C. § 4980H(a)(1). 125 More precisely, the I.R.C. §4980H(a) penalty is assessed based on the number of full-time employees excluding the first 30 employees. In a Request for Comments on Section 4980H, the Treasury Department hints that 4980H may be interpreted to require only that an employer offer coverage to “substantially all” of its full-time employees. No further clarification is given as to what is meant by “substantially all.” See http://www.irs.gov/pub/irs-drop/n-11-36.pdf at 18 (“It is contemplated that the proposed regulations would make clear that an employee offering coverage to all, or substantially all, of its full-time employees would not be subject to the §4980H(a) assessable payment provisions.”).



22

employer could offer very expensive insurance to lower-income employees in order to avoid the 4980H(a) penalty while still allowing those employees to qualify for the premium tax credits. As long as the lower-income employees’ required contributions for the health insurance would exceed 9.5 percent of the employees’ household incomes, the insurance would be considered “unaffordable” – making the employees eligible for the premium tax credits.

In contrast to the 4980H(a) penalty, the 4980H(b) penalty will be triggered when an employer does offer health insurance, but when that insurance is deemed “unaffordable.”126 Also unlike the 4980H(a) penalty, the 4980H(b) penalty is tailored to actual the number of employees who qualify for the Exchange subsidies (because the employer’s offer of health insurance was “unaffordable”), rather than being based on the total number of employees.127 More specifically, for any month in which it is triggered, the 4980H(b) penalty will equal one-twelfth of $3,000 multiplied by the number of full-time employees that receive the Exchange subsidies because the insurance offered to them by their employer was “unaffordable.”128 However, the amount of the 4980H(b) penalty will be limited so that it can never exceed the amount that the employer would have been liable for had the 4980H(a) penalty been triggered instead. Consequently, for any month, the 4980H(b) penalty is limited to a maximum of one-twelfth of $2,000 multiplied by the total number of full-time employees employed during the month.129

The 4980H(b) penalty will thus raise the cost to an employer of not offering “affordable” health insurance to lower-income employees so that those employees can qualify for the Exchange subsidies. The amount by which the 4980H(b) penalty will raise the costs of this strategy depends on the percentage of an employer’s workforce that ends up qualifying for the Exchange subsidies. An employer with only a small percentage of low- and moderate-income employees might face the full 4980H(b) penalty of $3,000 annually per employee that qualifies for the Exchange subsidies. In contrast, an employer with a large percentage of low- and moderate-income employees might have the 4980H(b) assessable payments limited to the 4980H(a) penalty amount of $2,000 annually multiplied by the total number of full-time employees.

The following charts, again adapted from a study by the Tax Policy Center,130 show the relative costs to employers of either offering employer-sponsored health insurance or sending

126 The 4980H(b) penalty may also be triggered by offers of affordable health insurance that fail the minimum value test of I.R.C. § 36B(c)(2)(C)(ii). 127 I.R.C. § 4980H(b)(1). 128 Id. 129 I.R.C. § 4980H(b)(2). More precisely, the penalty is assessed based on the number of full-time employees excluding the first 30 employees. 130 Rennane & Steurle, supra note __, at Tables 2 & 4. Note that the Tax Policy Center’s analysis underlying these charts is based on holding employers’ costs constant. I will not explain the Tax Policy Center’s methodology here, as interested readers can find that explanation in the Tax Policy Center’s report. Id.



23

employees to the Exchanges in 2016. The analysis incorporates the value of the Exchange subsidies, the tax exclusions, and the 4980H employer-mandate penalties. The analysis uses the 4980H(a) penalty amounts of $2,000 per employee annually, adjusted for inflation. The 4980H employer-mandate penalties would be higher, making the breakeven thresholds lower, for employers with only a small percentage of lower-income employees who would thus be subject to the 4980H(b) penalty of up to $3,000 annually per employee qualifying for the Exchange subsidies. The first chart shows the breakeven analysis for an employee with four members in the employee’s household and the second chart shows the breakeven analysis for an individual employee. For both charts, the right-most column (“Net Benefit of Exchange Coverage”) shows the total additional value that could be received from an employer not offering health insurance so that an employee can qualify for the Exchange subsidies. This “Net Benefit of Exchange Coverage” is equal to the value of the Exchange subsidies minus both the additional taxes that would be paid (e.g., from not taking advantage of the tax exclusions) and the employer-mandate penalty.

Employee's Household

Income as a % of FPL


Amounts

Total Value of Exchange

Subsidies

Total Increase in Taxes Paid

Section 4980H

Employer Penalty

Net Benefit of Exchange

Coverage100 24,000$ 18,433$ (123)$ 2,247$ 16,309$ 125 30,000$ 18,307$ 2,298$ 2,247$ 13,763$ 150 36,000$ 15,617$ 4,569$ 2,247$ 8,801$ 175 42,000$ 14,759$ 5,536$ 2,247$ 6,976$ 200 48,000$ 11,544$ 5,457$ 2,247$ 3,840$ 225 54,000$ 10,473$ 4,493$ 2,247$ 3,733$ 250 60,000$ 9,053$ 3,545$ 2,247$ 3,261$ 275 66,000$ 7,776$ 3,545$ 2,247$ 1,984$ 300 72,000$ 6,952$ 3,545$ 2,247$ 1,160$ 325 78,000$ 6,468$ 3,545$ 2,247$ 676$ 350 84,000$ 5,761$ 3,545$ 2,247$ (31)$ 375 90,000$ 5,165$ 3,545$ 2,247$ (627)$ 400 96,000$ 4,570$ 3,545$ 2,247$ (1,222)$ 425 102,000$ -$ 3,545$ 2,247$ (5,792)$ 450 108,000$ -$ 4,135$ 2,247$ (6,382)$

Breakeven Analysis for an Employee Receiving Exchange Coverage(Family of Four, 2016)



24

As the charts demonstrate, the primary impact of the employer-mandate penalties will be to lower the breakeven thresholds for the household-income levels at which employers and employees would jointly benefit from the employer not offering “affordable” health insurance so that the employee can qualify for the Exchange subsidies. Nevertheless, many low- and moderate-income employees will still benefit more from receiving the Exchange subsidies than from receiving employer-sponsored health insurance. For a family of four in 2016, the breakeven threshold for the household-income level at which it will be more cost effective for an employer to not offer “affordable” health insurance might be somewhere between 325 percent and 350 percent of the federal poverty line. For an individual employee, the breakeven threshold might be somewhere between 225 percent and 250 percent of the federal poverty line. The numerous employees with household incomes below the relevant breakeven thresholds would benefit more from their employers not offering “affordable” health insurance and paying the employer-mandate penalties so that the employees can qualify for the premium tax credits.

To recap, the 4980H(a) penalty will not prevent employers from providing health insurance to their higher-income employees while sending their lower-income employees to the Exchanges, because the employers can avoid the 4980H(a) penalty by offering their lower-income employees “unaffordable” health insurance. Only the 4980H(b) penalties will apply to an employer who offers “unaffordable” health insurance to its lower-income employees. Consequently, the primary effect of the 4980H employer-mandate penalties will be to lower the

Employee's Household

Income as a % of FPL


Amounts

Total Value of Exchange

Subsidies

Total Increase in Taxes Paid

Section 4980H

Employer Penalty

Net Benefit of Exchange

Coverage100 11,800$ 6,736$ 866$ 2,247$ 3,623$ 125 14,700$ 6,676$ 1,142$ 2,247$ 3,286$ 150 17,700$ 5,574$ 913$ 2,247$ 2,414$ 175 20,600$ 5,156$ 901$ 2,247$ 2,008$ 200 23,600$ 3,869$ 1,040$ 2,247$ 582$ 225 26,500$ 3,347$ 1,040$ 2,247$ 60$ 250 29,500$ 2,718$ 1,040$ 2,247$ (568)$ 275 32,400$ 2,095$ 1,040$ 2,247$ (1,192)$ 300 35,400$ 1,686$ 1,040$ 2,247$ (1,601)$ 325 38,300$ 1,452$ 1,040$ 2,247$ (1,834)$ 350 41,300$ 1,100$ 1,040$ 2,247$ (2,187)$ 375 44,200$ 812$ 1,040$ 2,247$ (2,475)$ 400 47,200$ 514$ 1,050$ 2,247$ (2,783)$ 425 50,100$ -$ 1,340$ 2,247$ (3,587)$ 450 53,100$ -$ 1,453$ 2,247$ (3,700)$

Breakeven Analysis for an Employee Receiving Exchange Coverage(Individual, 2016)



25

breakeven thresholds by raising the cost of not offering lower-income employees “affordable” health insurance by an amount equal to the 4980H(b) penalties.

Of course, employers will not be able to perfectly separate their lower-income employees (whose household incomes fall below the breakeven thresholds) from their higher-income employees (whose household incomes are above the breakeven thresholds). Employers will not always know their employees’ household incomes, as household income may derive from sources other than from the employer. Additionally, employers may find it difficult to vary the cost of the health insurance they offer to match the breakeven thresholds for employees with different family sizes. Nevertheless, the 4980H employer-mandate penalties will not prevent employers from exploiting the mismatch in available health care subsidies by simultaneously allowing many of their lower-income employees to qualify for the Exchange subsidies and many of their higher-income employees to benefit from the tax exclusions.

In contrast to the 4980H employer-mandate penalties, the nondiscrimination rules created by new Section 2176 of the Public Health Service Act may actually prevent employers from offering health insurance to their higher-income employees while allowing their lower-income employees to qualify for the Exchange subsidies. New Section 2716 of the Public Health Service Act is relatively short. In full, it reads:131

“SEC. 2716. PROHIBITION ON DISCRIMINATION IN FAVOR OF HIGHLY COMPENSATED INDIVIDUALS. (a) IN GENERAL.—A group health plan (other than a self-insured plan) shall satisfy the requirements of section 105(h)(2) of the Internal Revenue Code of 1986 (relating to prohibition on discrimination in favor of highly compensated individuals). (b) RULES AND DEFINITIONS.—For purposes of this section—

(1) CERTAIN RULES TO APPLY.—Rules similar to the rules contained in paragraphs (3), (4), and (8) of section 105(h) of such Code shall apply.

(2) HIGHLY COMPENSATED INDIVIDUAL.—The term ‘highly compensated individual’ has the meaning given such term by section 105(h)(5) of such Code.”

Section 105(h) of the Internal Revenue Code provides that self-funded health insurance plans cannot discriminate in favor of highly compensated individuals either with respect to eligibility to participate or with respect to the benefits provided under the plan.132 New Section

131 The Patient Protection and Affordable Care Act of 2010, H.R. 3590, 111th Congress, January 5, 2010, § 2716, pg. 766 – 767, available at http://democrats.senate.gov/pdfs/reform/patient-protection-affordable-care-act-as-passed.pdf; Karen Levin, INTERNAL REVENUE SERVICE, Notice 2010-63 – Request for Comments on Requirements Prohibiting Discrimination in Favor of Highly Compensated Individuals in Insured Group Health Plans, October 12, 2010, available at http://www.irs.gov/irb/2010-41_IRB/ar07.html. 132 I.R.C. § 105(h)(2).



26

2716 of the Public Health Service Act thus creates non-discrimination rules for insured plans “similar to” those that previously existed for self-insured plans.133

Prior to the ACA, the Section 105(h) nondiscrimination rules were only minimally enforced as employers were rarely audited for compliance with these rules.134 Recognizing that further guidance will be needed to clarify the new Section 2716 nondiscrimination rules, the Treasury Department and the IRS, along with the Department of Labor and HHS, “have determined that compliance with § 2716 should not be required (and thus, any sanctions for failure to comply do not apply) until after regulations or other administrative guidance of general applicability has been issued under § 2716.”135

Because such guidance has yet to be issued, I will not in this Essay analyze the extent to which the new nondiscrimination rules of Section 2716 are likely to be effective. For now, it is an open question as to whether and to what extent the new nondiscrimination rules might prevent employers from designing their health insurance offerings so that higher-income employees can take advantage of the tax exclusions while lower-income employees can qualify for the Exchange subsidies because they are only offered “unaffordable” health insurance. The new nondiscrimination rules may interfere with this strategy, at least to some extent, but employers may also find ways to effectuate this strategy while complying with the Section 2716 nondiscrimination rules.136

III) HOW THE ACA WILL CREATE EFFECTIVE TAXES

The previous two Parts explained how the ACA will alter employers’ incentives as to whether to offer health insurance. Crucially, the ACA will create a mismatch between the tax subsidies available for Exchange coverage and those available for employer-sponsored coverage, such that most lower-income taxpayers would receive more tax benefit from Exchange coverage whereas most higher-income taxpayers would receive more tax benefit from employer-sponsored coverage. By creating this mismatch in the available tax subsidies, and then attempting to nevertheless incentivize employers to offer affordable health insurance even for their lower-

133 Section 2716 only applies to insured plans that are not grandfathered. “However, it is anticipated that plan sponsors will find it difficult to maintain the grandfathered status of their plans. Therefore, as a practical matter, most, if not all, insured plans will eventually be subject to the new discrimination rule.” http://www.shearman.com/files/Publication/2f843bcc-ac80-49c2-8bed-7d4f50ded8b9/Presentation/PublicationAttachment/9f41c894-d516-480f-91e1-3a4b2286db95/ECEB-122110-Executive-Medical-Coverage-After-Health-Reform.pdf at 2. 134 E.g., id., at 1 (“How the discrimination rules under § 105(h)(2) of the Internal Revenue Code apply to medical plans has never been completely clear. However, many employers did not have to confront the ambiguities in the health plan discrimination rules under § 105(h)(2) because the rules applied only to self-insured arrangements and the rules were rarely enforced or ruled upon by the Internal Revenue Service.”) 135 http://www.irs.gov/pub/irs-drop/n-11-01.pdf at 3. 136 For further discussion, see Part III.B. infra.



27

income employees, the ACA will impose significant and harmful effective taxes with respect to low- and moderate-income workers.

Explaining how the ACA will create these effective taxes and analyzing their implications is the principal contribution of this Essay. Yet before proceeding it is useful to first briefly summarize the analysis from the previous two Parts.

A primary goal of the ACA is to fix the problems that have plagued the individual market for health insurance. However, a major reason why employers offered health insurance prior to the ACA was because the health insurance available on the individual market was inferior to what the employers were able to provide. Hence, to the extent the ACA succeeds in the (laudable) goal of fixing the individual market, the ACA will have removed a key driver behind employers offering health insurance.

The ACA retains the tax exclusions for employer-provided health insurance – the other key driver for why employers offered health insurance prior to the ACA. But for lower-income taxpayers, the value of these tax exclusions will be more than overpowered by the subsidies the ACA makes available for Exchange coverage. And taxpayers with offers of “affordable” employer-sponsored health insurance will be disqualified from receiving the major subsidies for Exchange coverage. The ACA will thus create a framework wherein lower-income taxpayers would benefit more from not being offered “affordable” employer-sponsored health insurance (so that they can qualify for the Exchange subsidies) whereas higher-income taxpayers would benefit more from receiving employer-provided health insurance (so that they can benefit from the tax exclusions). This mismatch between the tax subsidies for Exchange coverage and those for employer-sponsored coverage will create incentives for employers to offer “affordable” health insurance only to their higher-income employees.

The Obama Administration and the Congressional Democrats were very concerned about whether the ACA would lead employers to stop offering health insurance, both in order to contain the budgetary costs of the Exchange subsidies and because they deemed it politically important to assure voters that anyone who wanted to retain their previous health coverage would be able to do so. As a result, the ACA includes additional provisions designed to incentivize employers to continue offering health insurance. The most important of these provisions are the employer-mandate penalties and the new nondiscrimination rules.

The employer-mandate penalties will not prevent employers from providing “affordable” health insurance only to their higher-income employees. But the employer-mandate penalties will raise the costs of this strategy by an amount between $2,000 and $3,000 annually (adjusted for inflation) per full time-employee qualifying for the Exchange subsidies. In contrast, the new nondiscrimination rules may prevent employers from offering “affordable” health insurance only to their higher-income employees. But as regulatory guidance interpreting the new



28

nondiscrimination rules has yet to be released, it is difficult to predict whether the nondiscrimination rule will actually prevent this strategy.

If employers are prevented from offering “affordable” health insurance only to their higher-income employees, then employers may have to choose between either dropping health insurance for all of their employees (so as not to prevent their lower-income employees from receiving the Exchange subsidies) or offering “affordable” health insurance to all of their employees (so as not to deny their higher-income employees the benefits of the tax exclusion). Either choice would deny some employees the full value of the tax subsidies they might otherwise have taken advantage of. Moreover, either choice will create harmful effective taxes with respect to low- and moderate-income workers.

In contrast, to the extent employers are able to offer “affordable” health insurance only to their higher-income employees, employers will be able to facilitate both their lower-income and higher-income employees qualifying for the maximum possible health care subsidies. If employers are able to maximize their employees’ tax subsidies in this way, the budgetary cost of the Exchange subsidies may end up being much higher than predicted. Furthermore, employers will likely need to incur costs in order to offer “affordable” health insurance only to their higher-income employees. Employers might incur these costs as a result of being subject to the employer-mandate penalties or the nondiscrimination rules, or employers might incur these costs in order to implement strategies for avoiding being subject to those rules. In either case, these costs will create harmful effective taxes with respect to low- and moderate-income workers.

The effective taxes that the ACA will impose on low- and moderate-income workers can be grouped into two categories: (1) effective taxes created by the design of the Exchange subsidies, and (2) effective taxes created by the employer penalties. There is considerable overlap between these two categories of effective taxes, but it is nevertheless useful to consider the categories one at a time.

A) The Effective Taxes Created by the Design of the Exchange Subsidies

From an efficiency perspective, the impact of a tax is determined by the extent to which the tax alters the relevant prices of economic decisions.137 When a law raises the cost of one economic choice relative to alternative choices, this can be viewed as an effective tax on the choice that is made more expensive.138 The way in which the ACA structures its Exchange

137 For an explanation of how taxes impose efficiency costs – or, alternatively, “excess burden” or “deadweight loss” – see David Gamage & Darien Shanske, Three Essays on Tax Salience: Market Salience and Political Salience, 65

TAX LAW REVIEW 19, 61-64 (2011). 138 Sometimes the term “effective tax” is used more narrowly. E.g., Boris Bittker, Effective Tax Rates: Fact or Fancy?, 122 U. PA. L. REV. 780, 781 (1974) (“When the taxpayer's actual income tax liability is expressed as a fraction of a base other than taxable income, the resulting percentage is usually described as the ‘effective rate.’). But the term “effective taxes” can also be used broadly, as in this Essay. E.g., Daniel Shaviro, Effective Marginal Tax Rates on Low-Income Houses, 84 TAX NOTES 1191, 1192 (1999) (“My focus on effective marginal tax rates



29

subsidies will create effective taxes with respect to at least three important decisions: the choice to work for employers offering “affordable” health insurance, the choice to seek higher income, and the choice to marry (or, equivalently, the choice not to divorce).

The effective taxes that the ACA will impose on the decision to work for employers that offer “affordable” health insurance are probably among the most important of the effective taxes that this Essay will analyze. Most low- and moderate-income workers would receive considerably more value from the Exchange subsidies than from the tax exclusions. Yet if employers offer “affordable” health insurance to their low- and moderate-income workers, these workers will become disqualified from receiving the Exchange subsidies. The ACA will thus impose effective taxes on low- and moderate-income workers’ choice of whether to work for an employer offering “affordable” health insurance equal to the degree to which the tax exclusions would offer these workers less value than would the Exchange subsidies.139

For many low- and moderate-income workers, accepting a job with an employer that offers “affordable” health insurance will result in a net loss of thousands of dollars of health care subsidies.140 For instance, in 2016, a family of four with a household income of $60,000 would suffer a net loss of approximately $5,500 in health care subsidies from being offered “affordable” employer-sponsored health coverage.141 For a family of four with a household income of $36,000, the net loss of health care subsidies would be approximately $11,000 – almost a third of the family’s household income.142 These effective taxes will strongly deter low- and moderate-income taxpayers from accepting jobs that offer “affordable” health insurance.143

yields a very different perspective . . . . the analysis cannot stop with provisions that are formally denominated “taxes.” The reason one should care about marginal tax rates is that they show how government policy is affecting incentives and the distributional consequences of people’s decisions. . . . A marginal tax rate analysis thus must take account of all government programs that are either directly or indirectly income-conditioned.”). 139 As noted earlier, supra notes __ and accompanying text, if Exchange coverage remains inferior to employer-sponsored coverage even after the ACA, then the difference in value between the Exchange subsidies and the tax exclusions must be discounted by the degree to which Exchange coverage offers inferior value. 140 See the charts following note __ supra. 141 These calculations are derived from the breakeven charts based on the Tax Policy Center’s analysis. Id. These calculations factor out the employer-mandate penalties, because the goal here is to demonstrate the loss in health care subsidies to an employee from being offered affordable employer-sponsored health insurance; the goal is not to show breakeven analyses or employers’ incentives. 142 Id. This loss in health care subsidies would be mitigated for families able to qualify for other government health care programs. Notably, although a family of four with household income of $36,000 in 2016 should not qualify for Medicaid, as the family’s income would be too high, the children might still qualify for the Children’s Health Insurance Program (or CHIP). 143 Of course, the actual magnitude of the effective taxes on accepting jobs that offer affordable health insurance will depend on how much income the taxpayers could have earned from their alternative choices – from not working for employers that offer affordable health insurance. A full calculation of effective taxes would need to incorporate the many factors affecting the returns to work. The figures shown above are at best rough approximates, but they should still suffice to illustrate that the net loss of health care subsidies may be very large.



30

In addition to these effective taxes, the ACA will create further effective taxes because the value of the Exchange subsidies will phase out as household income increases.144 For instance, in 2016, a family of four without an offer of “affordable” employer-sponsored health insurance would lose approximately $6,500 in Exchange subsidies from increasing household income from $36,000 to $60,000.145 And for taxpayers who move from lower-paying jobs that do not offer “affordable” health insurance to higher-paying jobs that do offer “affordable” health insurance, these two forms of effective taxes will stack.146 Returning to our family of four in 2016, switching from earning a household income of $36,000 without “affordable” employer-sponsored health insurance to a household income of $60,000 with “affordable” employer-sponsored health insurance would result in a net loss of approximately $12,000 in health care subsidies.147

The ACA’s effective taxes on accepting jobs that offer “affordable” health coverage may be especially problematic because they affect the extensive margin of labor supply. The extensive margin of labor supply refers to the choice of whether or not to work; in contrast, the intensive margin refers to the choice of how many hours to work.148 A strong finding of the empirical economics literature is that labor-supply elasticities for low-income workers are much stronger along the extensive margin than along the intensive margin.149 Indeed, men appear to be “almost completely irresponsive” to effective taxes along the intensive margin, but “very responsive” to effective taxes along the extensive margin.150 Women are more responsive than men along both margins; but like men, women are even more responsive along the extensive margin than the intensive margin.151 Consequently, an effective tax operating on the extensive margin is far more likely to affect labor-supply decisions – and thereby reduce economic efficiency – than would a similarly sized tax affecting only the intensive margin.152

144 Seth Chandler has previously analyzed the effective taxes the ACA will create by phasing out the Exchange subsidies as household income increases. Seth J. Chandler, The Architecture of Contemporary Healthcare Reform and Effective Marginal Tax Rates 29 MISS. C. L. 335 (2010). Additionally, for a more general discussion of how phasing out social welfare benefits can generate high effective marginal tax rates, see Daniel Shaviro, Effective Marginal Tax Rates on Low-Income Households, 84 TAX NOTES 1191 (1999). 145 Id. 146 Moreover, these effective taxes created by the design of the Exchange subsidies will further stack with the effective taxes created by the design of other social welfare programs and with payroll taxes. See Chandler, supra note __; Shaviro, supra note __. 147 Id. 148 Emmanuel Saez, Optimal Income Transfer Programs: Intensive Versus Extensive Labor Supply Responses, 117 QUARTERLY J. OF ECON. 1039, 1039 (2002). 149 Id. at 1039-1040. 150 Costas Meghir & David Phillips, Labour Supply and Taxes, PREPARED FOR THE REPORT OF A COMMISSION ON REFORMING THE TAX SYSTEM FOR THE 21ST CENTURY, CHAIRED BY SIR JAMES MIRRLEES, at 44-45 (2008). 151 Saez, supra note __, at 1056. 152 For instance, a number of scholars have criticized the effective marginal taxes created by the phase out of tax benefits like the earned income tax credit (EITC). E.g., Shaviro, supra note __, at 1194. However, the manner in which the EITC phases out primarily affects only the intensive margin, not the extensive margin, such that the EITC may phase out in a close to optimal fashion. Saez, supra note __, at 1064-65 (“The combined EITC and U. S.



31

The ACA’s denial of the Exchange subsidies to taxpayers with offers of “affordable” employer-sponsored health insurance will create effective taxes along the extensive margin by reducing the net benefit low-income taxpayers would receive from accepting a job with an employer offering “affordable” health insurance.153 Moreover, there are reasons to think that when low-income workers decide against accepting formal-sector jobs this can generate negative externalities.154 As Nobel-Prize-winning economist Edmund Phelps has argued, when low-income taxpayers decide against formal-sector work, this decision can harm the taxpayers’ children, the taxpayers’ neighborhoods, and broader society, in addition to the taxpayers themselves.155 Hence, to the extent the ACA’s effective taxes discourage low-income taxpayers from accepting formal-sector employment, these effective taxes may prove especially harmful.

Because an offer of employer-sponsored health insurance will result in an employee’s entire family being disqualified from the Exchange subsidies156 – as long as the employee’s self-only coverage is “affordable”, and regardless of the cost of family coverage – the ACA will also impose effective taxes on marriage. Imagine a couple with children where one of the adults works for an employer that offers “affordable” health insurance. If the couple marries, then the non-employee spouse and children will be made ineligible for the Exchange subsidies. Similarly, if the couple is already married, then they can make the non-employee spouse and children qualify for the Exchange subsidies by divorcing. Hence, for many low-income families, the ACA will impose thousands of dollars in effective taxes on couples that choose to be legally married.

Despite the potential magnitude of these effective taxes, it may be the case that few couples will base their decision of whether to become legally married on the effective taxes.157 Nevertheless, even if the impact on behavior is not large, “marriage penalties” have historically had high political salience in tax law.158 There may well be political backlash once voters come

welfare system for single mothers is close to our optimal simulated schedules if, as evidenced by empirical studies, participation elasticities are substantial.”). In contrast, both the manner in which the Exchange subsidies phase out and the denial of the Exchange subsidies to employees with offers of affordable employer-sponsored health insurance affect the extensive margin of labor supply. 153 Note that the manner in which the effective taxes created by the design of the Exchange subsidies affects the extensive margin of labor supply is somewhat complicated by the interaction between the Exchange subsidies and Medicaid; however, exploring the implications of these interactions is beyond the scope of this Essay. 154 For a definition of negative externalities, see David Gamage, Taxing Political Donations: The Case for Corrective Taxes in Campaign Finance, 113 YALE L.J. 1283, 1292 (2004). 155 EDMUND S. PHELPS, REWARDING WORK: HOW TO RESTORE PARTICIPATION AND SELF-SUPPORT TO FREE

ENTERPRISE 38-48 (Harvard University Press, 1997). 156 I.R.C. § 36B(c)(2)(C)(i)(II); Prop. Treas. Reg. § 36B, 76 Fed. Reg. 50931, 50935 (Aug. 17, 2011). 157 But see, e.g., Kasey Buckles, Melanie Guldi, and Joseph Price, Changing the Price of Marriage: Evidence from Blood Test Requirements, at 24-25 (2009) (“We have shown that even small changes in the cost of marriage can have significant effects, particularly for certain populations.”), available at http://www.nd.edu/~kbuckles/BGP_nber.pdf. 158 See Lawrence Zelenak, Doing Something About Marriage Penalties: A Guide for the Perplexed, 54 TAX L. REV. 1 (2000) (discussing the political history of the debate over marriage penalties).



32

to understand the extent of the marriage penalties created by the ACA. And many commentators seem to agree that there is something inherently wrong with tax law imposing high marriage penalties even apart from efficiency concerns.159

The ACA’s effective taxes both on marriage and on accepting jobs that offer “affordable” health coverage could be considerably mitigated by basing the affordability of employer-sponsored health insurance for employees with families on the cost of family coverage, rather than on the cost of the employee’s self-only coverage. A number of commentators have suggested that the text of new Section 36B of the Internal Revenue Code is sufficiently ambiguous that the Treasury Department could interpret the affordability test for employees with families as being based on the cost of family coverage.160 The Treasury Department’s indication in its proposed regulations that the affordability test will be based on the cost of employee’s self-only coverage has thus been strongly criticized.161

I cannot at this time discuss my views regarding whether the Treasury Department would have the authority to interpret Section 36B so that the affordability test for employees with families would be based on the cost of family coverage. However, it is worth noting that the Treasury Department’s interpretation follows an earlier technical explanation by the Joint Committee on Taxation,162 and that the Congressional Budget Office’s official scoring of the cost of the ACA was also based on this interpretation.163 In any case, I can discuss the policy considerations underlying this question. If the affordability test were based on the cost of family coverage, rather than self-only coverage, then a secondary question would arise as to whether employers who offered “affordable” self-only coverage but “unaffordable” family coverage would be subject to the Section 4980H(b) employer-mandate penalties. 159 Leslie A. Whittington & James Alm, Marriage Penalty, NTA ENCYCLOPEDIA OF TAXATION AND TAX POLICY, available at http://www.taxpolicycenter.org/taxtopics/encyclopedia/Marriage-Penalty.cfm (“The principal arguments revolve around equity issues.”). 160 Peter Gosselin, New Rule Could Narrow Aid For Health-Plan Buyers and Shrink Insurers’ Sales, BLOOMBERG

GOVERNMENT STUDY, at 10 (2011) (“But the law is not quite as clear what type of coverage it intends the individual's contribution to go toward — coverage for just the individual or his or her family. And, to the extent the measure appears to apply the contribution to individual-only coverage, many observers thought the result was sufficiently unfair that Treasury and the IRS would use their regulation-writing authority to address the matter.”). 161 E.g., Id. at 20-21; Larry Levitt & Gary Claxton, Measuring the Affordability of Employer Health Coverage, KAISER FAMILY FOUNDATION, Aug 24, 2011 available at http://healthreform.kff.org/notes-on-health-insurance-and-reform/2011/august/measuring-the-affordability-of-employer-health-coverage.aspx?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+NotesOnHL+%28Notes+on+Health+Insurance+and+Reform+%28Headlines%29+-+Kaiser%27s+Health+Reform+Source%29, retrieved Sept. 18, 2011. 162 JOINT COMMITTEE ON TAXATION, GENERAL EXPLANATION OF TAX LEGISLATION ENACTED IN THE 111TH

CONGRESS, JCS–2–11 (March 2011) at 265 (‘‘Unaffordable is defined as coverage with a premium required to be paid by the employee that is more than 9.5 percent of the employee’s household income, based on the self-only coverage’’). See also Prop. Treas. Reg. § 36B, 76 Fed. Reg. 50931, 50935 (Aug. 17, 2011) (citing the Joint Committee on Taxation). 163 EMPLOYMENT POLICIES INSTITUTE, AN OFFER YOU CAN’T REFUSE: ESTIMATING THE COVERAGE EFFECTS OF THE

2010 AFFORDABLE CARE ACT, at 2 (2011), available at http://epionline.org/downloads/110715_EPI_AnOfferYouCantRefuse_PolicyBrief_Final.pdf.



33

If these employers would be subject to the Section 4980H(b) employer-mandate penalties, then employers would need to offer “affordable” family coverage in order to avoid those penalties. Arguably, it is not reasonable to expect the employers of low- and moderate-income workers to offer “affordable” family coverage, as this would dramatically increase the cost of hiring those workers. And rising health care costs would exacerbate this problem over time. Moreover, employers would face strong incentives to not hire employees with families to the extent they could legally do so. Perhaps most problematic, the Section 4980H(b) employer-mandate penalties are probably not high enough to incentivize employers to offer affordable family coverage.164 If a significant number of employers were to decide against offering affordable family coverage, or to drop offering coverage all together, then this could substantially increase the budgetary cost of the Exchange subsidies.165

Conversely, if employers offering “affordable” self-only coverage but unaffordable family coverage would not be subject to the Section 4980H(b) employer-mandate penalties, then employers would have strong incentives to make family coverage “unaffordable” for low- and moderate- income workers. Employers could shift any funds previously used to subsidize family coverage to instead increase the subsidies for self-only coverage. According to a study by the Employment Policies Institute, “it could be in the interest of a surprisingly large number of employees—and their employers—to change their current contracts so that otherwise-ineligible workers will be able to receive the exchange subsidy.”166 This would dramatically increase the budgetary cost of the Exchange subsidies.167

We thus return to the reason why the ACA denies the Exchange subsidies to employees with offers of “affordable” employer-sponsored health insurance. The drafters of the ACA wanted the Exchange subsidies to make insurance affordable for those who were previously uninsured, but also wanted to limit the budgetary cost of the Exchange subsidies by not making the subsidies available to those who previously received affordable health insurance from their employers. In order to prevent employers from dropping health insurance so that their employees could take advantage of the Exchange subsidies, the ACA thus also includes the employer-mandate penalties and the new nondiscrimination rules. It is this framework for limiting the budgetary cost of the Exchange subsidies that creates most of the effective taxes discussed in this Essay.

Basing the affordability test on the cost of family coverage would mitigate the effective taxes on marriage and on accepting jobs that offer “affordable” health insurance, but only by

164 And strengthening the employer-mandate penalties (which would require legislative action) would exacerbate another form of effective taxes on low- and moderate-income workers, as discussed in Part III.B infra. 165 Lacking offers of “affordable” employer-sponsored coverage, many more low-income workers and their families would qualify for the Exchange subsidies. 166 EMPLOYMENT POLICIES INSTITUTE, supra note __. 167 Id. at 4.



34

eroding the firewall that the ACA creates on the availability of the Exchange subsidies and thereby greatly increasing the budgetary cost of the Exchange subsidies.168 Conversely, the Treasury Department’s proposed rule of basing affordability on the cost of self-only coverage maintains the firewall and thereby contains the budgetary cost of the Exchange subsidies, but at the price of greatly exacerbating the ACA’s effective taxes. Moreover, as explained by the Employment Policy Institute, the Treasury Department’s proposed rule means that “millions of families will be stuck in a no-man’s-land without affordable coverage through their employer or the exchange – since family members of an employee with an offer of coverage are disqualified from accessing subsidized exchange coverage.”169

B) The Effective Taxes Created by the Employer Penalties

To the extent that employers decide to offer “affordable” health insurance to their low- and moderate-income employees, the ACA will create effective taxes primarily due to the design of the Exchange subsidies.170 Conversely, to the extent that employers decide against offering “affordable” health insurance to their low- and moderate-income employees, the ACA will create effective taxes primarily through the employer penalties.171 Both the new Section 4980H employer-mandate penalties and the new nondiscrimination rules will impose effective taxes on low- and moderate-income workers when employers decide against offering “affordable” health insurance.172 Moreover, these employer penalties will impose effective taxes regardless of whether employers actually end up paying the penalties or whether employers instead reorganize their business operations so as to avoid being subject to the penalties.

The effective taxes that will be created when employers are subject to the Section 4980H employer-mandate penalties are relatively straightforward.173 When employers become subject to the Section 4980H(b) employer-mandate penalties (of between $2,000 and $3,000 annually,

168 Basing the affordability test on the cost of family coverage would not necessarily mitigate the effective taxes on increasing household income, as these effective taxes result from the Exchange subsidies phasing out as household income increases rather than from the Exchange subsidies being unavailable to taxpayers with offers of affordable employer-sponsored health insurance. 169 Id. 170 Of course, to the extent that the employer penalties are what cause employers to offer affordable health insurance to their low- and moderate-income employees, it would be more accurate to say that the effective taxes are caused by the interaction between the design of the Exchange subsidies and the employer penalties. The effective taxes I analyze in Part III.A are thus not entirely distinct from the effective taxes I discuss in Part III.B. In many respects, effective taxes impacting employees are equivalent to effective taxes impacting employers, with the ultimate result depending on the incidence of the effective taxes. Nevertheless, despite the overlap between the effective taxes discussed in Part III.A and Part III.B, I have separated the two discussions for ease of exposition. 171 The effective taxes created by the Exchange subsidies phasing out as household income increases will also apply when employers do not offer affordable health coverage. See notes ___ supra and accompanying text. 172 These employer penalties also impose effective taxes when employers do offer affordable health insurance, due to the design of the Exchange subsidies, but these effective taxes were discussed in Part III.A. supra. 173 As discussed in Part II.C., employers should generally be able to avoid paying the Section 4980H(a) employer-mandate penalties by offering unaffordable health insurance to their low- and moderate-income employees. Consequently, the Section 4980H(a) employer-mandate penalties should not create significant effective taxes.



35

adjusted for inflation), the penalties will raise the cost of hiring employees who qualify for the Exchange subsidies. Economists generally agree that raising the cost to employers of hiring low- and moderate-income workers will lead employers to reduce those workers’ salaries, hire fewer low- and moderate-income workers, or implement some combination of these two coping strategies.174 The Section 4980H(b) employer-mandate penalties will thus function as an effective tax on employing low- and moderate-income workers.175

The new nondiscrimination rules may similarly function as an effective tax on employing low- and moderate-income workers, depending on whether these rules succeed in preventing employers from offering “affordable” health insurance only to their higher-income employees. The penalties for violating the new nondiscrimination rules are extremely severe – at $100 per day per affected employee.176 Consequently, employers should not purposefully allow themselves to become subject to the penalties for violating the new nondiscrimination rules. Instead, employers should either offer health insurance in a nondiscriminatory fashion or else stop offering health insurance altogether.

To the extent that the new nondiscrimination rules motivate employers to stop offering health insurance even to their higher-income employees, these employers’ low- and moderate-income employees will be able to qualify for the Exchange subsidies. But because these employers will be unable to offer their higher income employees the benefit of the tax exclusions, the new nondiscrimination rules will raise the costs to these employers of offering attractive compensation packages to their higher-income employees. These increased costs will function as effective taxes in that the employers will need to make adjustments in order to offset for the increased costs. Some portion of the incidence of these effective taxes will affect the employers’ hiring of low- and moderate-income employees (while another portion will affect the employers’ hiring of higher-income employees). Hence, in addition to the effective taxes the new Section 4980H employer-mandate penalties will impose on hiring low- and moderate-income employees, the new nondiscrimination rules will impose additional effective taxes to the extent that the nondiscrimination rules interfere with the employers’ offering attractive compensation packages to their higher-income employees.

Rather than allowing themselves to be subject to the new Section 4980H employer-mandate penalties or the new nondiscrimination rules, many employers will reorganize their business operations so as to qualify for exceptions to these rules. In particular, employers may

174 See, e.g., Monahan & Schwarcz, supra note __, at 43 (“This is because most economists agree that health care costs are simply part of employees’ total compensation, and thus that decreased health insurance costs will tend to translate into increased salaries.”); Lawrence Summers, Some Simple Economics of Mandated Health Benefits, 79 AM. ECON. REV. 177 (1989). See also note __ supra. 175 For further discussion, see, e.g., John Goodman, The $6-an-Hour Health Minimum Wage, blog post, October 18, 2010, available at http://healthblog.ncpa.org/the-6-an-hour-min-wage/. 176 PHSA §§ 2722 & 2716.



36

shift some of their low- and moderate-income workers to part-time positions.177 Neither the Section 4980H employer-mandate penalties nor the new nondiscrimination rules apply to part-time employees, with part-time defined as employees working an average of less than 30 hours per week with respect to the Section 4980H employer-mandate penalties178 and as employees working less than 35 hours per week with respect to the nondiscrimination rules.179 By moving their low- and moderate-income employees to part-time status, employers can avoid having to either offer these employees “affordable” health insurance or be subject to the employer-mandate penalties or the non-discrimination rules.

Moreover, low- and moderate-income employees should often prefer to work part-time and not be offered “affordable” health insurance than to work full time while being offered “affordable” health insurance.180 Shifting their low- and moderate-income employees to part-time status can thus facilitate employers maximizing the value of the health care tax benefits for all employees, as lower income employees would be eligible for the Exchange subsidies and higher-income employees for the tax exclusions – all without the employer being subject to either the Section 4980H employer-mandate penalties or the new nondiscrimination rules.

Moving low- and moderate-income workers to part-time status is probably among the most promising strategies employers might use to reorganize their business operations so as to avoid the employer-mandate penalties and the nondiscrimination rules. But many alternative strategies may also be available. For instance, employers might replace some of their employees with independent contractors or move their low- and moderate-income employees into separate business divisions from their higher-income employees.181 With respect to the nondiscrimination rules, it is not yet clear whether offering “unaffordable” insurance to lower-income employees would even be considered discriminatory if the insurance is offered on similar terms to all employees.182 Until final guidance is released interpreting both Section 4980H and the new nondiscrimination rules, it would be premature to attempt a full analysis of possible avoidance

177 See, e.g., Diana Furchtgott-Roth, Job Creation and the Affordable Care Act, 132 TAX NOTES 1289, 1289 (2011)

(“That combination of taxes gives businesses a powerful incentive to downsize, replace full-time employees with part timers, and contract work out to other firms or individuals.”). 178 I.R.C. § 4980H(c)(4). 179 PHSA § 2716 provides that group health plans (other than self-insured plans) must satisfy rules “similar to” the I.R.C. § 105(h) rules governing self-insured plans. Because full guidance interpreting PHSA § 2716 has yet to be released, there is uncertainty on how the “similar to” language will be applied. That caveat aside, Treas. Reg. § 1.105-11(c)(2) specifies that part-time employees (generally defined as employees working less than 35 hours per week) are excluded from the 105(h) nondiscrimination rules. 180 Notes ___ supra. 181 E.g., Diana Furchtgott-Roth, supra note __, at 1289-90; Eugene Steuerle, Health Care Reform: Implications of a Two Subsidy System, PRESENTATION AT THE AEI CONFERENCE ON FAIRNESS IN HEALTH REFORM, at 13 (December 4, 2009), available at http://www.urban.org/uploadedpdf/509103_healthcarereform.pdf. 182 Insurance offered on similar terms to all employees could end up being unaffordable to lower-income employees due to the simple fact that lower-income employees have less income with which to purchase insurance.



37

strategies; nevertheless, many high-priced law firms and consulting firms are undoubtedly already hard at work designing these strategies.183

To the extent employers can effectively implement avoidance strategies and thereby offer “affordable” health insurance only to their higher-income employees, the budgetary cost of the Exchange subsidies may be much higher than predicted. The purpose of the Section 4980H employer-mandate penalties and the new nondiscrimination rules is to contain the cost of the Exchange subsidies by denying those subsidies to employees with offers of “affordable” employer-sponsored health insurance. The Congressional Budget Office (CBO) scored the budgetary cost of the ACA based on the assumption that most employees who had offers of “affordable” employer-sponsored health insurance prior to the ACA will continue to have those offers after the ACA comes into effect.184 The CBO’s estimates appear to assume that few employers will be able to use avoidance techniques so as to offer “affordable” health insurance only to their higher-income employees.185 If employers can successfully offer “affordable” health insurance only to their higher-income employees, than a much larger number of taxpayers will be eligible for the Exchange subsidies than the CBO predicted, greatly increasing the budgetary cost of the Exchange subsidies.186

Moreover, employers will need to incur costs in order to reorganize their business operations. For instance, if an employer relied on full-time workers prior to the ACA, this implies that hiring full-time workers made more sense from a business perspective as compared to hiring part-time workers. If the ACA induces the employer to switch to hiring part-time

183 For the most recent update on the status of the proposed guidance interpreting I.R.C. §4980H, see Notice 2012-17, Frequently-Asked-Questions From Employers Regarding Automatic Enrollment, Employer Shared Responsibility, And Waiting Periods, available at www.dol.gov/ebsa/faqs/faq-aca5.html . 184 See CBO’S ANALYSIS OF THE MAJOR HEALTH CARE LEGISLATION ENACTED IN MARCH 2010, TESTIMONY

BEFORE THE SUBCOMMITTEE ON HEALTH, COMMITTEE ON ENERGY AND COMMERCE, U.S. HOUSE OF

REPRESENTATIVES, STATEMENT OF DOUGLAS W. ELMENDORF, DIRECTOR, at 19-21 (March 30, 2011) (discussing the CBO’s estimates for the number of employers dropping health coverage and stating “[s]ome commentators have expressed surprise that CBO and JCT do not expect a much larger reduction in employment-based insurance coverage….”). 185 See Id. at 20-21 (arguing that employers will not stop offering affordable coverage to their lower-income employees because the employers will not want to forgo offering their higher-income employees the value of the tax exclusions and stating that the nondiscrimination rules will prevent employers from doing both.). 186 The cost of the Exchange subsidies may greatly exceed the CBO’s estimates even if employers are able to avoid only the nondiscrimination rules and not also the § 4980H employer-mandate penalties. As discussed in Part II.C supra, many low-income employees would receive more net benefit from the Exchange subsidies than from being offered affordable employer-sponsored insurance even when employers are subject to the § 4980H penalties. But if employers can avoid being subject to both the § 4980H penalties and the nondiscrimination rules, employers will have incentives to stop offering affordable coverage for a much larger number of low- and moderate-income employees. On the other hand, if the ACA fails in fixing the problems of the individual market such that Exchange coverage remains significantly inferior to employer-sponsored coverage, then employers will face much less incentive to stop offering affordable health insurance to their low- and moderate-income employees. Yet few supporters of the ACA are likely to be comforted by the idea of costs being contained through the ACA failing in one of its primary goals.



38

workers in order to avoid the employer-mandate penalties or the nondiscrimination rules, this will create costs for the employer to the extent that hiring part-time workers would otherwise make less business sense than would hiring full-time workers.

Most any strategy employers might use to reorganize their business operations to avoid the employer-mandate penalties or the nondiscrimination rules will create costs for the employer. And at least some of the incidence of these costs will fall on low- and moderate-income workers – to the extent the costs lead employers to reduce hiring or to reduce salaries. And even to the extent the incidence of these costs falls on higher-income workers (or on owners, managers, or customers), this will still create economic harm.187

Overall then, the ACA will impose harmful effective taxes on low- and moderate-income workers regardless of how employers respond to the ACA’s framework. These effective taxes may be somewhat less severe if employers stop offering “affordable” health insurance to their low- and moderate-income employees. But then the cost of the Exchange subsidies will likely be much higher than predicted. Although it is not yet clear which forms of effective taxes will end up creating the most harm – partially because full regulatory guidance interpreting the employer-mandate penalties and nondiscrimination rules has yet to be released – there is little doubt that both employees and employers will face strong perverse incentives once key provisions of the ACA come into effect in 2014.

IV) CONCLUSION: IMPLICATIONS FOR HEALTHCARE REFORM

Although this Essay has focused on critiquing aspects of the ACA, this Essay should not be interpreted as a criticism of the ACA as a whole.188 If the ACA succeeds in fixing the individual market for health insurance, that will be an amazing accomplishment.189 And the

187 Because the penalty savings that induce employers to incur these costs represent money lost to the government, the costs create pure losses from a social welfare perspective. 188 Note __ supra.

In my view, assessing the ACA as a whole largely boils down to the importance one places on providing affordable health insurance options to those with pre-existing conditions. Any healthcare reform designed to provide insurance to those with pre-existing conditions must either rely on government provided health care (i.e., “socialized medicine”) or else on a hybrid system similar to the ACA. Because I believe it important to provide health insurance options to those with pre-existing conditions, I conclude that the ACA’s benefits will exceed its costs, even though I wish that the ACA were better designed so as to avoid creating unnecessary costs such as those discussed in this Essay. But most analysts who do not place a high importance on preventing insurance companies from discriminating against those with pre-existing conditions (or otherwise providing affordable insurance options to those with pre-existing conditions) are likely to conclude that the ACA’s costs exceed its benefits. 189 It is by no means clear whether the ACA will succeed in fixing the individual market for health insurance (even ignoring the threat of constitutional or political challenges). For papers discussing some of the issues that will need to be resolved in order for the ACA to be successful in this endeavor, see, e.g., Amy Monahan & Daniel Schwarcz, Will Employers Undermine Health Care Reform by Dumping Sick Employees?, 97 VIRGINIA L. REV. 125 (2011); Pamela Short et al., Realizing Health Reform’s Potential: Maintaining Coverage, Affordability, and Shared



39

ACA includes many provisions designed to slow the growth of health care costs, to expand health care coverage to the previously uninsured, and to achieve other laudable goals.190 That the ACA will also impose harmful effective taxes on low- and moderate-income workers does not mean that these workers would be better off without the ACA. After all, most of the effective taxes that the ACA will impose on these workers will only arise to the extent that the ACA successfully improves the health insurance available on the individual market and offers generous subsidies to make that insurance “affordable” to low- and moderate-income taxpayers.191

But neither should the ACA’s positive achievements distract from our understanding the harms that the ACA will impose through effective taxes. To the extent that the ACA successfully incentives employers to continue offering “affordable” health insurance to their low- and moderate-income employees, the ACA will strongly discourage low- and moderate-income taxpayers from marrying or staying married and from accepting jobs that offer “affordable” health insurance. Conversely, if numerous employers stop offering “affordable” health insurance to their low- and moderate-income employees, then the budgetary costs of the Exchange subsidies will likely greatly exceed the CBO’s projections and the ACA will disincentive employers from hiring low- and moderate-income workers. To summarize, however employers respond to the ACA’s framework, the ACA’s perverse incentives will reduce employment opportunities for low- and moderate-income Americans as well as creating other economic and social harms.

If these perverse incentives were unavoidable, then perhaps we should view them as a necessary cost of achieving the ACA’s desirable ends. Yet the ACA’s laudable goals could have been attained without imposing most of the harmful effective taxes discussed in this Essay.

The mismatch between the tax subsidies available for Exchange coverage and those available for employer-sponsored coverage are the source of most of the perverse incentives that

Responsibility When Income and Employment Change, THE COMMONWEALTH FUND 1503 (May 2011); Benjamin D. Sommers & Sara Rosenbaum, Issues in Health Reform: How Changes in Elgibility May Move Millions Back and Forth Between Medicaid and Insurance Exchanges, 30 HEALTH AFFAIRS 228 (2011); Timothy Jost, Health Insurance Exchanges and the Affordable Care Act: Eight Difficult Issues, THE COMMONWEALTH FUND 1444

(September 2010); Sarah Lueck, State Should Structure Insurance Exchanges to Minimize Adverse Selection, CENTER ON BUDGET AND POLICY PRIORITIES (August 17, 2010). 190 There are many explanations of the various provisions of the ACA. My favorite is a video produced by the Kaiser Family Foundation, available at http://healthreform.kff.org/the-animation.aspx. 191 As noted previously, most of the effective taxes discussed in this Essay only arise to the extent that the ACA succeeds in fixing the individual market so that the insurance available on the individual market is competitive with employer-sponsored coverage. See, e.g., notes __ and accompanying text supra. Throughout most of this Essay, I have assumed that the ACA will succeed in fixing the individual market. But this assumption is made for ease of exposition. It remains to be seen whether (and to what extent) the ACA will improve the individual market to the degree necessary to make the insurance offered on the Exchanges competitive with employer-provided offerings. See note __ supra.



40

the ACA will create with respect to low- and moderate-income workers.192 Yet there is a simple fix for this mismatch. The ACA could have been drafted to replace the tax exclusions for employer-provided health insurance with refundable tax credits structured similarly to the Exchange subsidies.

Ideally, the ACA would have replaced the tax exclusions for employer-sponsored health insurance with refundable tax credits offering the same value as the Exchange subsidies.193 If – for any given level of household income – a taxpayer would obtain the exact same subsidies for either employer-sponsored coverage or for Exchange coverage, then there would be no perverse incentives affecting either marriage or accepting jobs that offer “affordable” health insurance. Moreover, there would be no need for the employer-mandate penalties or the nondiscrimination rules, as employers would not face any perverse incentives to stop offering health coverage to their low- and moderate-income employees. Instead, Exchange coverage and employer-sponsored coverage could compete on a level playing field, with neither benefitting from a larger tax advantage.194

Many previous commentators have called for replacing the tax exclusions with refundable tax credits on the grounds that the tax exclusions are regressive and that they

192 The major effective taxes that do not arise from the mismatch between the tax subsidies for employer-sponsored coverage and those for Exchange coverage (or from the ACA’s provisions designed to incentivize employers to continue offering affordable health insurance to low- and moderate-income employees despite this mismatch) are the effective taxes that result from how the Exchange subsidies phase out as household income increases. See Part III.A. This Essay’s argument for replacing the tax exclusions with refundable tax credits for employer-sponsored coverage designed in a similar fashion to the Exchange subsidies would thus not eliminate the effective taxes created by the manner in which the subsidies phase out as household income increases. To eliminate these effective taxes, the tax subsidies available both on and off of the Exchanges would need to be redesigned so that they do not phase out with income, in effect making them available for high income taxpayers as well as for low- and moderate-income taxpayers (like a demogrant). Analyzing the tradeoffs involved in addressing the effective taxes created by the manner in which the tax subsidies phase out as household income increases is beyond the scope of this Essay. For further discussion, see Chandler, supra note __. 193 If repealing the tax exclusions would not generate sufficient revenue to make tax credits of similar magnitude to the Exchange subsidies available to all low- and moderate-income workers with employer-sponsored coverage, then I would argue that some of the funding for the Exchange subsidies should be moved to funding the new credits for employer-sponsored coverage so that equivalent health care subsidies would be available both on and off the Exchanges. There might be some justifications for subsidizing Exchange coverage more than employer-sponsored coverage, as bringing a critical mass of insureds to Exchange coverage is important for combating adverse selection. But there does not appear to be any good reason for incentivizing only low- and moderate-income taxpayers to receive Exchange coverage while incentivizing higher-income taxpayers to receive employer-sponsored coverage, as the ACA will do as it is currently structured. Even if we cannot completely eliminate the effective taxes that the ACA will impose on low- and moderate-income workers, we should strive to mitigate these effective taxes, by equalizing the tax subsidies available on and off of the Exchanges to the extent possible. 194 A question underlying much of healthcare policy is whether the American system of relying primarily on employer-sponsored coverage should be maintained or eroded. This Essay takes no position on this question. Instead, this Essay’s suggested reforms would permit Exchange coverage and employer-sponsored coverage to compete on an even playing field, without either enjoying a significantly greater tax advantage. Whether Exchange coverage would replace employer-sponsored coverage, then, would depend on which form of coverage could offer better value at lower cost.



41

encourage excess health care consumption.195 This Essay’s key contribution is to show that failing to replace the tax exclusions with refundable tax credits will create far more harm once key provisions of the ACA go into effect in 2014. In addition to being regressive and encouraging excess health care consumption, retaining the tax exclusions in their current form will impose harmful effective taxes on low- and moderate-income workers.

Fortunately, there is some cause for hope. Many Republicans have already embraced the idea of replacing the tax exclusions with refundable tax credits as part of their vision for reforming health care. Replacing the tax exclusions with refundable tax credits was a centerpiece of Senator McCain’s healthcare plan when he ran for President in 2008.196 And, more recently, Republican House Budget Committee Chairman Paul Ryan has argued for replacing the tax exclusions with refundable tax credits as part of his “Path to Prosperity” plan.197 If Democrats can be convinced of the importance of replacing the tax exclusions with refundable tax credits in order to avoid the harmful effective taxes that the ACA will otherwise impose on low- and moderate-income workers, then there may be room for a bipartisan compromise based on the Republicans’ proposals. Even if such a compromise is not possible within the current hyper-partisan atmosphere surrounding health care discussions in Washington,198 there may still be hope after President Obama leaves office and the vitriolic debate over the enactment of the ACA recedes into more distant memory.

Although the ACA is a landmark accomplishment,199 the task of reforming the American healthcare system is far from complete.200 Rising health care costs will force politicians to revisit healthcare reform in the not-too-distant future.201 If the problem of the effective taxes that

195 E.g, Goldberg & Camic, supra note __; Holahan et. al., Containing the Growth of Spending in the U.S. Health System, URBAN INSTITUTE HEALTH POLICY CENTER, at 11-13 (October 2011); Jonathan Gruber, The Tax Exclusion for Employer-Sponsored Health Insurance, NBER WORKING PAPER 16766, at 7-8 (2010). 196 Goldberg & Camic, supra note __ at 3; Lucinda E. Jesson, Beyond Efficiency: Creating Health Policy Through the Tax Code, at 21 (forthcoming, 2010), available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1695673. 197 Avik Roy, Paul Ryan, in a Major Speech, Proposes Universal Health Coverage Via Tax Credits, FORBES, September 28th, 2011, available at http://www.forbes.com/sites/aroy/2011/09/28/paul-ryan-in-a-major-speech-proposes-universal-health-coverage-via-tax-credits/. 198 It is worth noting that key Republican legislators have made it clear that the House will not pass any measures designed to improve the Affordable Care Act, even if the measures would otherwise have been noncontroversial or would have been supported by Republicans. At the time of this writing, many Republican legislators would apparently prefer for the ACA to be as ineffective and harmful as they can make it, so that the Republicans can score political points against Democrats. Hopefully, Republicans will be more open to improving the ACA in the future, after further election cycles have passed. 199 STAFF OF THE WASHINGTON POST, LANDMARK: THE INSIDE STORY OF AMERICA’S NEW HEALTH CARE LAW AND

WHAT IT MEANS FOR US ALL (PublicAffairs, 1st ed., 2010). 200 E.g., Health-Care Reform is an Ongoing Process, THE ECONOMIST, Jan 23, 2011, available at http://www.economist.com/blogs/democracyinamerica/2011/01/health-care_reform (“Which is all to say that reform is an ongoing process.”). 201 See, e.g., David I. Auerbach & Arthur L. Kellermann, A Decade of Health Care Cost Growth Has Wiped Out Real Income Gains for an Average Family, 30 HEALTH AFFAIRS 1630 (2011), available at http://content.healthaffairs.org/content/30/9/1630.full.html (evaluating the burdens imposed by rising health care costs).



42

the ACA will impose on low- and moderate-income workers cannot be resolved sooner, then hopefully future rounds of healthcare reform can avoid the ACA’s mistakes. It may be politically advantageous to make explicit subsidies as high as possible and to keep explicit taxes as low as possible.202 But if this results in creating harmful effective taxes on important decisions affecting low- and moderate-income workers, then the political advantages are not worth the costs. In future healthcare debates, independent-minded commentators should focus on the danger of creating effective taxes that harm low- and moderate-income workers. These effective taxes will only be less visible to the extent they are not brought into the light through op-eds and other media discussions. By holding future politicians’ feet to the fire, we can strive for healthcare reform that will not create perverse incentives harming low- and moderate-income Americans.

202 See, e.g., Katherine Pratt, Funding Health Care With An Employer Mandate: Efficiency and Equity Concerns, 39 ST. LOUIS U. L.J. 155, 161 (1994) (“Many Americans would not view an employer mandate as a tax increase, even though they will ultimately bear the economic burden of the mandate.”). For a broad review of the literature related to the political salience of taxation and of government financing, see David Gamage & Darien Shanske, Three Essays on Tax Salience: Market Salience and Political Salience, 65 TAX LAW REVIEW 19, 33-54 (2011). In particular, see the discussion of “tax-financed spending versus regulation” in Part II.B.6.c., id.

Date post:	29-Aug-2020
Category:	Documents
Upload:	others
View:	7 times
Download:	0 times

gamage - how the ACA will create perverse …...The Affordable Care Act’s Perverse Incentives Tax...

Documents